API MPMS C19.4 97

Manual of Petroleum Measurement Standards Chapter 19.4-Recommended

Practice for Speciation of Evaporative Losses

FIRST EDITION, NOVEMBER 1997

Reaffirmed 3/2002

American Petroleum Institute

Copyright American Petroleum Institute Provided by IHS under license with API

Not for ResaleNo reproduction or networking permitted without license from IHS

--`,,-`-`,,`,,`,`,,`---

Manual of Petroleum Measurement Standards Chapter 19.4-Recommended

Practice for Speciation of Evaporative Losses

Measurement Coordination

FIRST EDITION, NOVEMBER 1997




--`,,-`-`,,`,,`,`,,`---

API publications necessarily address problems of a general nature. With respect to particular circumstances, local, state, and federal laws and regulations should be reviewed.

API is not undertaking to meet the duties of employers, manufacturers, or suppliers to warn and properly train and equip their employees, and others exposed, concerning health and safety risks and precautions, nor undertaking their obligations under local, state, or federal laws.

Information concerning safety and health risks and proper precautions with respect to particular materials and conditions should be obtained from the employer, the manufacturer or supplier of that material, or the material safety data sheet.

Nothing contained in any API publication is to be construed as granting any right, by implication or otherwise, for the manufacture, sale, or use of any method, apparatus, or product covered by letters patent. Neither should anything contained in the publication be construed as insuring anyone against liability for infringement of letters patent.

Generally, API standards are reviewed and revised, reaffirmed, or withdrawn at least every five years. Sometimes a one-time extension of up to two years will be added to this review cycle. This publication will no longer be in effect five years after its publication date as an operative API standard or, where an extension has been granted, upon republication. Status of the publication can be ascertained from the API Authoring Department [telephone (202) 682-8000]. A catalog of API publications and materials is published annually and updated quarterly by API, 1220 L Street, N.W., Washington, D.C. 20005.

This document was produced under API standardization procedures that ensure appropriate notification and participation in the developmental process and is designated as an API standard. Questions concerning the interpretation of the content of this standard or com- ments and questions concerning the procedures under which this standard was developed should be directed in writing to the director of the Authoring Department (shown on the title page of this document), American Petroleum Institute, 1220 L Street, N.W., Washington, D.C. 20005. Requests for permission to reproduce or translate all or any part of the material published herein should also be addressed to the director.

API standards are published to facilitate the broad availability of proven, sound engineering and operating practices. These Standards are not intended to obviate the need for apply- ing sound engineering judgment regarding when and where these standards should be utilized. The formulation and publication of API standards is not intended in any way to inhibit anyone from using any other practices.

Any manufacturer marking equipment or materials in conformance with the marking requirements of an API standard is solely responsible for complying with all the applicable requirements of that standard. API does not represent, warrant, or guarantee that such products do in fact conform to the applicable API standard.

__-

All rights reserved No part of this work muy be reproduced, stored in a retrieval system, or transmitted by any means, electronic, mechanical, photocopying, recording, or otherwise,

without prior written permission from the publishel: Contact the Publisher; API Publishing Services, 1220 L Street, N. W, Washington, D.C. 20005.

Copyright Q 1997 American Petroleum Institute



--`,,-`-`,,`,,`,`,,`---

FOREWORD

This publication contains recommended methods for estimating the yearly speciated organic compound emissions from multi-component hydrocarbon liquids stored in various types of tanks and from marine vessel transfer operations. These methods were developed for use with petroleum operations and are applicable to hydrocarbon mixtures such as crude oils, fuel oils, and gasoline.

In addition to the speciation methods, this document provides an overview of the emission sources and their respective mechanisms, as well as an overview of vapor-liquid equilibrium. A summary of key product related variables needed for speciation of the emissions from these sources is provided, along with detailed examples showing the speciation calculations.

API publications may be used by anyone desiring to do so. Every effort has been made by the Institute to assure the accuracy and reliability of the data contained in them; however, the Institute makes no representation, warranty, or guarantee in connection with this publication and hereby expressly disclaims any liability or responsibility for loss or damage resulting from its use or for the violation of any federal, state, or municipal regulation with which this publication may conflict.

Suggested revisions are invited and should be submitted to the Measurement Coordinator, American Petroleum Institute, 1220 L Street, N.W., Washington, D.C. 20005.

iii



--`,,-`-`,,`,,`,`,,`---

CONTENTS

Page

1 SCOPE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.1 Purpose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.2 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.3 Calculation Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.4 Document Organization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

2 REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 2.1 Referenced Standards. Codes. and Publications .......................... 2 2.2 Other References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

.

3 SYMBOLS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

4 KEYVARIABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 4.2 Stock True Vapor Pressure. P . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 4.3 Atmospheric Pressure. Pa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 4.4 Component Saturated Vapor Pressure. Pp .............................. 3 4.5 Stock Liquid Molecular Weight. M ,. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 4.6 Stock Vapor Molecular Weight. M , ..................................... 3 4.7 Component Molecular Weight. Mi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 4.8 Component Molar Concentration in the Liquid. xi . . . . . . . . . . . ._ . . . . . . . . . . . 3 4.9 Component Molar Concentration in the Vapor. yi ......................... 3 4.10 Stock Liquid Bulk Temperature. Tb . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 4.11 Typical Petroleum Product and Crude Values ............................ 3

5 TYPICAL CALCULATION EXAMPLE ................................... 4

6 EMISSION SOURCES AND LOSS MECHANISMS ......................... 5 6.1 Tank and Transfer Emission Sources .................................. 5 6.2 LossMechanisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

7 DESCRIPTION OF METHODOLOGIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 7.2 Level 1 (Available Representative Liquid Sample Analysis) . . . . . . . . . . . . . . . . 8 7.3 Level 2-Vapor Profile Available .................................... 13 7.4 Level 3-Vapor Sampk Collected ................................... 13 7.5 Common Mistakes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

8 EXAMPLE APPLICATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Example 1: Fixed-Roof Tank . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Example 2: Internal Floating-Roof Tank ............................... 16 Example 3: External Floating-Roof Tank .............................. 17 Example 4: Marine Vessel Transfer Operations ......................... 19

8.1 8.2 8.3 8.4

APPENDIX A-VALIDITY OF RAOULT’S LAW ............................. 21 APPENDIX B-DETERMINATION OF PHYSICAL PROPERTY DATA FOR SPECIFIC CHEMICALS AND PETROLEUM STOCKS ......................... 27

V



--`,,-`-`,,`,,`,`,,`---

Page

APPENDIX C-COMPARISON OF MOLECULAR WEIGHT. BOILING POINTS. VAPOR PRESSURE. AND BLENDING RVP FOR SELECTED HYDROCARBONS AND OXYGENATES ..................................................... 33 APPENDIX D-REFERENCES ............................................. 37

Figures C-1 Normal Boiling Point Versus Molecular Weight .......................... 34 C-2 Pure Substance Vapor Pressure at 100°F Versus Molecular Weight . . . . . . . . . . . 35

Tables

2 3 4

5 6 7

Weight Fractions ................................................... 11 8 Vapor Profile for Simulated Gasoline ................................... 13 9 Fixed Roof Tank Speciated Emissions for a U.S. Mid-Continent

Crude Oil ......................................................... 15 10 Internal Floating-Roof Tank Speciated Emissions for a

Simulated Gasoline ................................................. 16 11 External Floating-Roof Tank Speciated Emissions for a

Simulated Gasoline ................................................. 18 12 Marine Vessel Transfer Operations Speciated Emissions for a

Simulated Gasoline ................................................. 19 A-1 Summary of Results for Summer Blend Unleaded Gasoline ................. 21 A-2 Summary of Results for Winter Blend Unleaded Gasoline .................. 21 A-3 Compounds Selected for Speciation .................................... 21 A 4 GC Analysis Concentrations Using Average Response Factors (ARF) and

Linear Regression (LR) for Liquid Phase Samples (Concentrations in &fi) ........................................................... 22

A-5 GC Analysis Concentrations Using Average Response Factors (ARF) and Linear Regression (LR) for Vapor Phase Samples (Concentrations in pg/mL) ........................................................... 22

A-6 Comparison of Predicted Vapor Concentrations Using the Response Factor Analytical Data .................................................... 23

A-7 Comparison of Predicted Vapor Concentrations Using the Linear Regression Analytical Data .................................................... 24

B-1 Average Annual Stock Liquid Bulk Storage Temperature (Tb) as a Function of Tank Paint Color ................................................... 27

B-2 Vapor Pressures of Selected Compounds ................................ 30 B-3 Physical Properties and Antoine Constants for Selected Hydrocarbons ........ 31 C-1 Chemical Substances Found as Constituents in Crude Oil Fractions .......... 33

1 Symbols ........................................................... 2 sp ica l Petroleum Product and Crude Values ............................. 4 Typical Concentration. wt%. in Petroleum Stocks .......................... 4 Storage Tank Vapor Composition Example HAP in Vapor for 9.2 psi RVP Motor Gasoline Based on Raoult’s Law .................................. 4 Conversion of Liquid Concentrations to Liquid Mole Fractions ............... 9 Mid-Continent Crude Oil Vapor Phase Speciation Example ................. 10 Calculation of Partial Pressures. Vapor Mole Fractions. and Vapor

vi



--`,,-`-`,,`,,`,`,,`---

Recommended Practice for Speciation of Evaporative Losses

1 Scope This publication contains recommended methods for esti-

mating specific organic compound emissions from storage tanks, and marine vessel transfer operations handling multicomponent hydrocarbon mixtures (such as crude oils and gasoline) associated with petroleum operations. This publication was developed by the API Committee on Evaporation Loss Measurement.

This document assumes that the user has access to API Publication MPMS 19.2, on Floating-Roof Tanks; API Publi- cation MPMS 19.1, on Fixed-Roof Tanks; and API Publica- tion 25 14A, Marine Vessels, and will use those documents to calculate total hydrocarbon emissions. The methods in these documents are used to estimate annual losses from various types of tank construction including fixed and floating roofs, rim-seal systems and roof fittings, as well as varioiis liquid stocks, stock vapor pressures, tank sizes, and wind speeds. They are also applicable for estimating annual losses from loading and unloading operations for various types of marine vessels, types and volumes of cargo, and compartment treat- ment.

The methodology contained in this document is applicable for properly maintained equipment under normal working conditions. The approach was developed for speciating vapor emissions from liquids with vapor pressure that has reached equilibrium with ambient conditions (i.e., not boiling), stocks with a true vapor pressure less than 14.7 pounds per square inch absolute (psia), for average wind speeds ranging from 2 to 5 miles per hour, and for tank diameters greater than 20 feet.

1.1 PURPOSE

It should be noted that quality of the vapor speciation calculation is directly dependent on the liquid composition profile approximation. Due to averaging effects, calculations based on standard profiles for a large number of sources achieve better accuracy. As the number of sources decreases and as the time frame is dropped to less than one year these calculations have the potential for less accuracy.

The approach is not intended to be used for the following applications:

a. To estimate losses from unstable or boiling stocks. b. To estimate losses from hydrocarbons or petrochemicals for which the vapor pressure of some constituents is unknown or cannot be readily predicted. c. To estimate losses from tanks in which the materials used in the rim seal, roof fittings, or both, have either deteriorated or been significantly permeated by the stored stock.

1

1.2 METHODOLOGY

Methodologies for speciating emissions from petroleum tankage or marine operations are provided at three levels of detail and accuracy. The three levels are as follows:

a. Level 1-Calculation of hydrocarbon composition of vapor based on analysis of a representative stock liquid sample or representative stock liquid profile. Speciated emissions are calculated from analysis of liquid stock samples or representative profiles of liquid stock using an appropriate vapor- liquid equilibrium relationship. b. Level 2-Calculation of the hydrocarbon composition of vapor based upon representative vapor profiles. Speciated emissions are calculated using standard vapor profiles developed for representative stocks and can be used for similar stocks or blends. c. Level 3-Direct measurement of hydrocarbon composition of the vapor. Speciated emissions are calculated directly from analyses of vapor samples collected from the stock liquid in the storage tank.

1.3 CALCULATION PROCEDURE The general procedure for determining evaporative emis-

sions of specific hydrocarbons for the emission sources listed above is presented in Figure 1. Calculations of the hydrocarbon composition can either be made for emissions of the total vapor space involved, including the air in the vapor space (based on atmospheric pressure as the total system pressure), or on the hydrocarbon portion of the vapor space (based on the total hydrocarbon partial pressure of the vapor space as the system pressure). The best approach depends on how the results are to be used. These two approaches are discussed in detail in the sections that follow.

1.4 DOCUMENT ORGANIZATION

Section 4 provides a summary of the key product related variables needed for speciating hydrocarbon emissions from these sources. These variables are further discussed in Appendix B.

Section 5 provides a summary of the steps used in a typical calculation, and illustrates how the calculations could be set up in a spreadsheet program.

Section 6 provides an overview of the emission sources and the emission mechanisms from these sources.

Section 7 provides an overview of vapor-liquid equilibrium, and outlines the methodology for determining the gas- phase composition from liquid analyses. These concepts are further discussed in Appendix A.

Section 8 uses examples derived from API Publications 2514A, MPMS 19.1 and MPMS 19.2 as the basis for present- ing the methodology to illustrate speciation.



--`,,-`-`,,`,,`,`,,`---

2 CHAPTER 19.4

2 References 3 Symbols

2.1 REFERENCED STANDARDS, CODES, AND

API

Symbols used in this recommended practice are described PUBLICATIONS in Table 1.

4 Key Variables Pub1 25 14A Atmospheric Hydrocarbon Emission From Marine Vessel Transfer Operations

MPMS 19.1 Evaporation Loss From Internal Fixed Roof Tanks

MPMS 19.2 Evaporation Loss From Floating-Roof Tanks

4.1 INTRODUCTION

Information regarding the physical property and composition of the petroleum stock is required to accurately calculate total hydrocarbon emissions and the emissions of the individual hydrocarbon species. This information is discussed in Appendix B, and summarized below:

a. Stock true vapor PressUre, p- b. Component saturated vapor pressure, P,".

2.2 OTHER REFERENCES

listed in Appendix D. Footnoted references found throughout this publication are

Table 1 -Symbols

Symbol Description Units ~

Activity coefficient of component i

Total mass emissions

Mass emissions of component i

Henry's Law Constant of component i

Vapor-liquid equilibrium constant of component i

Molecular weight of component i

Stock liquid molecular weight

Stock vapor molecular weight

Stock true vapor pressure at bulk liquid temperature

Atmospheric pressure

Partial pressure of component i in the vapor

Saturated vapor pressure of pure component i

Total system pressure

Stock Reid vapor pressure

Stock ASTM-D86 distillation slope at 10 volume percent evaporated

Stock liquid temperature

Ambient temperature

Daily ambient temperature range

Stock liquid bulk temperature

Weight concentration of component i in the liquid

Density o Ï stock liquid

Density of condensed stock vapor

Molar concentration of component i in the liquid

Molar concentration of component i in the vapor

Weight concentration of component i in the vapor

Total weight concentration of hydrocarbons in the vapor

Antoine Constant in Equation B-3



dimensionless

Iblyr

Iblyr

psia

dimensionless

Ib/lb-mole

Ibllb-mole

Ibflb-mole

psia

psia

psia

psia

psia

psi

"FIvol.%

"F O F

"F O F

weight fraction

Iblgal

Iblgal

mole fraction

mole fraction

weight fraction

weight fraction



--`,,-`-`,,`,,`,`,,`---

RECOMMENDED PRACTICE FOR SPECIATION OF EVAPORATIVE LOSSES 3

c. Stock liquid molecular weight, M,. d. Stock vapor molecular weight, M,, e. Component molecular weight, Mi. f. Component molar concentration in the liquid, xi. g. Component molar concentration in the vapor, y;. h. Stock liquid storage temperature, Tb.

4.2 STOCKTRUE VAPOR PRESSURE, P For refined petroleum and crude oil stocks, the stock true

vapor pressure, P, of the stored stock liquid (in pounds psia) may be determined from the average Reid vapor pressure (RVP), and the average liquid bulk temperature, Tb, using the vapor pressure Equations B- 1 or B-2 shown in Appendix B. Some typical product and crude values are shown in the table at the end of this section. P is needed to determine total hydrocarbon emissions from thë emission source. In most situations, it will be assumed that the average bulk liquid temperature and the average stock true vapor pressure ( W P ) are essentially the same as those at the surface of the liquid where vaporization takes place.

4.3 ATMOSPHERIC PRESSURE, Pa The average atmospheric pressure, Pa, at the facility loca-

tion (in pounds psia) can be used to determine emissions in the total pressure calculation method. The atmospheric pressure may be either measured at the site, or a value of 14.7 pounds psia may be used.

4.4 COMPONENT SATURATED VAPOR PRESSURE, Pp

The specific vapor pressure, P,", of each component in the stock liquid at the stock temperature is needed to determine the stock vapor composition. This value can be obtained from the API Technical Data Book-Petroleum Refining, or can be calculated using the Antoine Constants, as outlined in Appen- dix B. Vapor pressures for a number of key chemicals and petroleum constituents are provided in Table B-2 (along with other physical data), and Antoine Constants are listed in Table B-3. The chosen temperature should be the same as that used in 2.2 for the stock WP.

4.5 STOCK LIQUID MOLECULAR WEIGHT, Ml The stock liquid molecular weight, M,, is the average molec-

ular weight of the liquid on a weight basis. The M, is needed to convert stock liquid concentrations from a weight basis to a molar basis, and can be determined from analysis of stock liquid, or calculated from the composition of the stock liquid. Two analytical methods are available: gel permeation chromatography (GPC) using a refractive index (RI) detector; and gas chromatography (GC) using a flame ionization detector (FiD). More information on these procedures can be found in [14].

4.6 STOCK VAPOR MOLECULAR WEIGHT, M, The stock vapor molecular weight, M,,, is also needed to

convert concentrations from a molar basis to a weight basis. It

can be determined either by analysis of vapor samples, or by calculation from the vapor composition. In the absence of this information, information from the table at the end of this section may be used, or a typical value of 64 IbAb-mole may be assumed for gasoline, and a value of 50 Ib/lb-mole may be assumed for U.S. mid-continent crude oils.

4.7 COMPONENT MOLECULAR WEIGHT, Ml The molecular weight of the individual components is

needed to convert concentrations from a mole basis to a weight basis. The molecular weights of a number of hydrocarbon compounds found in petroleum stocks are listed in Table B-2.

4.8 COMPONENT MOLAR CONCENTRATION IN THE LIQUID, Xi

The component molar concentrations of specific chemical constituents in the stock liquid, xi, are needed to calculate equilibrium vapor Concentrations. These liquid compositions can be determined from a representative summary of liquid stock samples, or from data on the stock liquids used to blend the petroleum product. These concentrations are sometimes presented on a volume (vi) or weight basis, wi, while the vapor-liquid equilibrium relationship is calculated on a molar basis, xi. The conversion from a volume (vi) or weight basis, wi, to a molar basis, xi, is shown in Appendix B.5.1.

4.9 COMPONENT MOLAR CONCENTRATION IN THEVAPOR, V,

When using an equation of state to determine the stock vapor composition, the component concentration in the vapor, y;, is determined on a volume or molar basis. To determine the quantity of VOC emissions represented by a specific constituent, the component concentration has to be converted to a weight basis, wi. This conversion is shown in Appendix B.5.2.

4.10 STOCK LIQUID BULKTEMPERATURE, ïb To estimate the vapor pressure of the individual components

in the petroleum stock and the composition of the vapor space, one must know the average liquid bulk temperature, T,. This can be determined from gauging records, or estimated from the Average Annual Ambient Temperature, T,, using the relationship shown in Appendix B, Table B-l. Normally, it is assumed that the liquid bulk temperature and the liquid surface temperature are the same within the precision of the overall calculations.

4.11 TYPICAL PETROLEUM PRODUCT AND CRUDE VALUES

Typical RVP, TVP, density and molecular weight values are based on data submitted to EPA and API during 1992 and 1993 in response to the Clean Air Act Section 1 14 Question- naires and are shown in Table 2.

The number of data point sets ranged as follows: gasoline, 132-234; crude, 34-106; JP-4,6--25; JPA, 24-95; and diesel, 39-163.



--`,,-`-`,,`,,`,`,,`---

4 CHAPTER 19.4

Table 2-Typical Petroleum Product and Crude Values

Vapor Pressure, psia Molecular Weight, Ibs/lbs-mole

RVP TVP Density, Liquid, Vapor, Petroleum Stock psia (13 70°F Ibdgal M , M”

9.2 6.5 6.0 94 67 Gasoline, Year Avg. Crude Oil 4.4 3.4 7.1 207 62 JP-4 2.3 1.2 6.2 120 99 JPA, Turbo, Kero 0.6 o. 1 6.8 160 130 Diesel 0.2 0.06 7.1 I85 130

Typical volatile organic hazardous pollutant (VOHAP) and other constituent data were taken from the same sources as above and are shown in Table 3.

The number of data point sets ranged as follows: gasoline, 49-237; crude, 22-79; JP-4, 5-27; JPA, 19-75; and diesel, 29-82.

5 Typical Calculation Example This section provides a typical calculation example for

individuals who are familiar with the principles of using Raoult’s Law, but might need a template to refresh their memories on the approaches to use for most situations. Sec- tions 4 and 5 provide background information to familiarize individuals with the fundamentals, or to evaluate non-routine situations.

This example uses a floating-roof tank, storing 9.2 psia RVP motor gasoline that is low in oxygenate composition.

Table &Typical Concentration, Wt%, in Petroleum Stocks

Typical Concentration, Wt%, in Petroleum Stocks

Constituent Gasoline Crude JP-4 JPA, Kero Diesel MTBE 3-7 O O O O Hexane 3-4 1.4 5 0.4 o. 1 n-Hexane 1 0.4 1-2 o. 1 0.04 Benzene 1.8 0.6 0.6 0.07 0.2 Iso-octane 2-6 <0.1 O O O Toluene 7 1 .o 1-3 0.3 0.4 Ethylbenzene 1.4 0.4 0.5 0.2 0.2 Xylenes 7 1.4 2-3 0.9 0.8 Cumene 0.5 o. 1 0.2 0.07 0.1

Because this example intends to estimate the amount of hazardous air pollutants emitted, the approach will be based on using the vapor pressure of the total hydrocarbons in the vapor space, and not the total pressure including air. Only a small amount of oxygenates are present, so it will be suffi- ciently accurate to use Raoult’s Law to estimate the component vapor compositions from their liquid compositions and normal vapor pressures. Raoult’s Law states that the partial pressure of a component in the vapor state is equal to the liquid’s molar fraction multiplied by its vapor pressure at the temperature of the liquid. The fractional composition of the liquid’s hydrocarbon vapor is determined by dividing its vapor partial pressure by the total hydrocarbon vapor pressure. Typically, liquid compositions are given as volume or weight percents. Therefore, most of the calculations involve

Table &Storage Tank Vapor Composition Example HAP in Vapor for 9.2 psi RVP Motor Gasoline Based on Raoult’s Law

Liquid Composition Vapor Composition

TVP Density Partial Pressure Gasoline Constituents at 75°F psia Mol Wt Iblgal Vol % W t % Mol% psia Mol% Wt% MTBE 4.7 88 6.2 1 1 1.1 0.05 0.8 1.1 n-Hexane 2.8 86 5.1 1.1 I 1.1 0.03 0.5 0.6 Benzene 1.7 78 7.4 I .5 1.8 2.2 0.04 0.6 0.7 Iso-octane 1 114 5.8 4.1 4 3.3 0.03 0.5 0.9 Toulene 0.5 92 7.3 5.8 I 7.2 0.04 0.6 0.8 Ethylbenzene 0.18 106 7.3 1.2 1.4 1.2 0.002 O O Xylenes 0.15 106 1.3 5.8 7 6.2 0.01 0.2 0.3 Cumene 0.08 120 7.2 0.4 0.5 0.4 0.0003 O O

Other - - - 79.1 76.3 77.3 6.3 96.8 95.6

Motor Gasoline Liquid 6.5 94 6 100 100 100 Vapor 67 6.5 100 100

Note: Liquid weight % = volume % times density of a constituent divided by density of gasoline; liquid mole % = weight ’?6 times molecular weight of gasoline divided by molecular weight of a constituent; vapor partial pressure of a constituent = its normal vapor pressure times its mol % in liquid; mol % in the hydrocarbon vapor = a constituent’s partial pressure divided by the total hydrocarbon partial pressure; the mol % in the total vapor space would be the constituent partial pressure divided by atmosphere pressure; weight % in the hydrocarbon vapor = a constituent’s mol % times its mol weight divided by the molecular weight of the hydrocarbon vapor.



--`,,-`-`,,`,,`,`,,`---


converting volume or weight percents to mol percents and back again. This is most easily accomplished by usinga spreadsheet in Lotus 1-2-3TM, ExcelTM, or a comparable program that can be replicated for multiple cases just by entering new liquid composition data. For tanks with products containing oxygenates, see 6.2.4.2 and Appendix C.

6 Emission Sources and Loss Mechanisms

This section describes the basic designs of storage tanks included in API Publications 2517, 2519, 19.1 and 19.2 [2,3,4, 251; marine vessel transfer operations included in API Publication 2514A [i]; and also discusses the loss mechanisms that occur from each of these sources.

6.1 TANK ANDTRANSFER EMISSION SOURCES

6.1.1 Fixed-Roof Tanks

A typical fixed-roof tank consists of a cylindrical steel shell, with a permanently affixed roof that may vary in design from cone- or dome-shaped to flat. The design of fixed-roof tanks requires an opening to the atmosphere that allows for movement of displaced air and vapors during filling, withdrawal, and expansion due to warming. The openings are commonly equipped with pressure/vacuum devices that allow the vessel to operate at a slight internal pressure or vacuum to prevent the release of vapors during very small changes in temperature, pressure, or liquid level. Of current tank designs, the fixed-roof tank is the least expensive to construct, and is generally considered to be the minimum acceptable equip ment for storage of organic liquids. Breathing losses and working losses are two significant types of emissions from fixed-roof tanks.

Breathing losses result from expulsion of stock vapor from a tank due to vapor expansion and contraction caused by changes in temperature and barometric pressure. When the barometric pressure rises, or the tank temperature drops, air is drawn into the tank vapor space and tends to become saturated with stock vapor. Upon subsequent expansion of the vapor, the vapor is expelled from the tank. This loss occurs even if no liquid stock is loaded or withdrawn from the tank.

Evaporative losses result from a change in liquid level in the tank and are called working losses. Working losses include both filling losses and emptying losses. Filling losses occur when the liquid levei in the tank increases and the pressure inside the tank exceeds the relief valve pressure setting; consequently, vapors are expelled. Emptying losses occur when air, drawn into the tank during liquid removal, becomes saturated with stock vapor, and then may be expelled from the tank prior to refilling.

Fixed-roof emissions vary as a function of tank capacity, vapor pressure of the stored liquid, throughput of the tank, and atmospheric conditions at the tank location. Emissions

from fixed-roof tanks can be reduced by installing internal floating roofs or by using add-on control devices such as vapor recovery or thermal oxidation. For additional information concerning fixed-roof tanks, see API MPMS 19.1 [3].

6.1.2 External Floating-Roof Tanks

External floating-roof tanks are cylindrical tanks with a roof that floats on the surface of the liquid being stored. The basic components of the tank include: (1) a cylindrical shell, (2) a floating roof, (3) an annular rim seal attached to the perimeter of the floating roof, and (4) roof fittings that penetrate the floating roof and serve operational functions. The purpose of the floating roof and the rim seal (or rim seal system) is to reduce the evaporative loss of the stored liquid. The liquid surface is completely covered by the floating roof except at the small annular space between the floating roof and the tank shell. A rim seal (or rim seal system) attached to the floating roof contacts the tank shell (with small gaps) and covers the annular rim space. The rim seal slides against the tank shell as the floating roof is raised or lowered. External floating-roof tanks have numerous roof fittings that pass through, or are attached to, the floating roof to allow for operational functions. The most common roof fittings that are sources of evaporative losses include: access hatches, guide pole wells, guide pole/sample wells, gauge float wells, gauge hatchísample wells, vacuum breakers, roof drains, roof legs, and rim vents. For more information on these roof fittings see , M I ,- Publication 2517 [2].

Emissions from external floating-roof tanks are the sum of the standing storage loss and the withdrawal loss. The standing storage loss can be estimated as the sum of the rim seal loss and the roof fitting loss. The rim seal loss occurs from evaporation of stock liquid past the primary and/or secondary seals. The roof fitting loss occurs from openings in the floating roof that expose the product or vapor, or both. Withdrawal losses occur as the liquid that clings to the tank shell is exposed to the atmosphere. This clingage then vaporizes when the floating roof is lowered by withdrawal of the stored liquid. Although the withdrawal loss is typically smaller than the standing storage loss, frequent turnovers of liquid in an external floating-roof tank can increase the withdrawal loss.

6.1.3 Internal Floating-Roof Tanks

An internal floating-roof tank has both a permanent fixed roof and an internai floating deck. The fixed roof reduces the wind speed to near zero; the internal deck reduces contact between the liquid surface and the tank vapor space, thereby reducing the evaporation of the stored liquid. The deck rises and falls with the liquid level, and either floats directly on the liquid surface (contact deck), or rests on pontoons several inches above the liquid surface (non-contact deck). Two basic types Òf internal floating-roof tanks are: (1) tanks with a fixed roof that is supported by vertical columns within the tank, and



--`,,-`-`,,`,,`,`,,`---

6 CHAPTER 19.4

(2) tanks with a self-supported fixed roof that has no internal support columns. Fixed-roof tanks that have been retrofitted with a floating deck are typically of the first type, while external floating-roof tanks retrofitted with a floating deck typically have a self-supported roof. Tanks initially constructed with both a fixed roof and a floating deck could be either of these types. Both types of decks incorporate rim seals that slide against the tank shell as the deck moves up and down. Floating roofs can have primary and secondary rim seals. Generally, these tanks are freely vented by circulation vents both at the top of the fixed roof and at the top of the shell. The vents minimize the possibility of stock vapor accumulation in concentrations approaching the flammable range. For additional information concerning internal floating-roof tanks, refer to APi Publication 25 19 [4].

The loss from internal floating-roof tanks is the sum of the withdrawal loss and the standing storage loss. The withdrawal loss from internal floating-roof tanks involves vaporization of stock liquid that clings to the tank shell and columns. The standing storage loss includes the rim seal loss, the deck fitting loss, and the deck seam loss. The rim seal loss involves evaporative loss from the rim seals. The deck fitting loss occurs from penetrations in the floating roof by deck fittings, fixed roof column supports, or other openings. The deck seam loss occurs from floating roofs that are bolted. Welded floating roofs have welded deck seams; consequently, there is no seam loss. Emissions from internal floating-roof tanks can be reduced by using liquid mounted primary seals, continuous rim mounted secondary seals, welded decks, and gasket fittings.

6.1.4 Marine Vessel Transfer Operations

Petroleum stocks are often transferred from one location to another ir. marine vessels such as barges, ships or tankers. These vessels usually contain a number of compartments to hold the petroleum stock. Because these compartments are not completely filled with stock, some vapor space is present. During transport, petroleum stock will evaporate into the vapor space. As the petroleum stock is loaded onto a marine vessel, some of these vapors are displaced and vented from the vessel. After the stock has been unloaded, vapors will remain in the compartment until additional petroleum stock or ballast water is added to the vessel. Ballast water is usually added at the unloading location, causing displaced vapors to exit to the atmosphere. When new stock is transferred into the vessel compartment, these vapors are also displaced, along with the new vapors that evaporate from the stock entering the compartment. Additional information concerning marine vessel transfer operations can be found in API Publication 2514A [i].

6.2 LOSS MECHANISMS The mechanisms by which hydrocarbons are released from

storage tanks and marine vessel transfer operations [9] were mentioned briefly in the prior section. These loss mechanisms are discussed in more detail below.

6.2.1 Breathing Loss

The vapor pressure of the stored liquid is one of the most important parameters affecting storage emissions. Vapor pressure is a function of temperature, barometric pressure, and product composition. External factors such as changes in temperature and barometric pressure, or mechanical agitation can cause the liquid to evaporate. During evaporation, most liquids tend to establish an equilibrium concentration of vapors above the liquid surface. Under completely static conditions, an equilibrium vapor concentration would be estab- lished, and no further evaporation would occur. At this point, the tank vapor space is saturated with vapor.

Breathing losses from the vapor space occur when daily temperature and barometric pressure changes cause thermal expansion and contraction of the vapor, This causes saturated vapor to be expelled from the tank and fresh air to be drawn in. Typically, volatile liquids with a true vapor pressure exceeding 1.5 psia are no longer stored in fixed-roof tanks in the US. A methodology to estimate emissions from low vapor pressure substances stored in tanks has been developed. The method utilizes a standing storage loss equation which was developed from a theoretical model of the breathing loss process. Equations for breathing losses are presented in API MPMS 19.1 [3]. In addition, factors affecting breathing losses include the tank diameter, average outage, and tank color (heat absorption).

6.2.2 Working Loss

The working loss is the evaporative loss associated with a change of liquid level in the tank, and can include both the displacement of vapor by a rising liquid surface, as well as the venting of vapor following a rapid stock liquid withdrawal. It is also dependent upon the number of turnovers that result from the volume of liquid that is pumped through a tank.

Again, an important factor is the stock liquid true-vapor pressure. Working losses are also dependent upon the volume of liquid being pumped into the tank.

6.2.3 Rim Seal Loss

The mechanism of vapor loss from the rim seals of floating-roof tanks is complex. However, wind has been found to be the dominant factor in inducing rim-seal vapor losses from external floating-roof tanks. Wind tunnel tests discussed in API 2517 [2] indicate that as the wind flows up and over the top of an external floating-roof tank, it produces a low-pressure zone above the floating roof on the upwind side of the



--`,,-`-`,,`,,`,`,,`---


tank. This results in air from the downwind side of the floating roof moving around its circumference above the rim seal. Therefore, a steady wind establishes pressure differences across a floating roof, with higher pressures on the downwind side, and lower pressures on the upwind side. This pressure difference establishes an airflow across the floating roof and induces losses in two basic ways. In one case, the pressure differences cause air to enter any continuous rim vapor space beneath the rim seal on the downwind side of the floating roof, which then flows circumferentially through the rim vapor space, flushing an air-hydrocarbon mixture out past the rim seal on the upwind side of the floating roof. This convec- tion reduces the hydrocarbon concentration in the rim vapor space, causing more liquid to evaporate to reestablish equilibrium conditions. The magnitude of this wind-induced loss depends on the tightness of the rim seal system and the pres- ence of any gaps between the rim seal and the tank shell.

When no continuous rim vapor space exists beneath the rim seal and above the liquid, the above condition does not occur. In this case, the wind flowing above the rim seal produces turbulence in the air that is present in any gaps between the rim seal and the tank shell. This turbulence causes fresh air to mix with the hydrocarbon vapor within the gap, and results in a reduction in the hydrocarbon concentration within the gap. This causes more liquid to evaporate to reestablish equilibrium conditions.

Other potential causes of rim seal loss include the expansion of gas in the rim vapor space, which is attributable to changes in temperature or pressure (breathing); permeation-of the sealing fabric by vapor; and wicking of the liquid in the rim space up the tank shell into contact with the air above the rim seal. However, studies conducted by the API indicate that losses from these mechanisms are typically negligible.

Rim seal losses vary with the true vapor pressure of the stock and the average molecular weight of the vapor, as well as with the average wind speed, and the rim seal loss factor.

6.2.4 Total Roof Fitting Loss

The same mechanisms that affect the rim seal loss also affect losses through fittings in the floating roof. These fittings, which penetrate the floating roof, are potential sources of loss because they require openings that allow for commu- nication between the stored liquid and the vapor space above the floating roof. While such openings are routinely sealed, the design details of certain roof fittings generally preclude the use of a completely vapor-tight seal.

The total roof fitting loss varies with the stock liquid true- vapor pressure, the diameter of the tank, and the individual roof fitting loss factors, which are dependent upon the design of the individual roof fitting.

6.2.5 Withdrawal Loss

As a floating roof descends during stock withdrawal, some of the liquid stock clings to the inside surface of the tank shell and is exposed to the atmosphere. Evaporation losses represent the extent to which this clingage evaporates before it is covered by the ascending floating roof during a subsequent filling. Generally, the most important product factor affecting stock clingage is the viscosity of the stock liquid, with refined stocks having lower clingage than crude oil. Other factors affecting the withdrawal loss include the tank diameter, the stock liquid density, and the shell condition (light rust, dense rust, or Gunite lining).

The withdrawal loss can generally be considered negligible, relative to the standing storage loss, except in cases where the tanks have a high degree of throughput. Unlike the other types of evaporative loss, the withdrawal loss is characterized by the composition of the stock liquid rather than the stock vapor.

6.2.6 Deck Seam Loss

For internal floating-roof tanks, emissions can also occur from the deck seams. It is assumed that this loss either occurs continuously, or from discrete localized points that are dis- tributed along the entire length of the deck seam. Again, an important factor is the stock vapor pressure. The deck seam loss is also affected by the tank diameter, the deck seam length and the seam loss factor. The deck seam loss factor is zero for welded decks.

6.2.7 Transfer Loss

In ships, barges, and tankers, petroleum stock evaporates into the compartment vapor space, eventually reaching equilibrium if no vapor is displaced. When the ship or barge is loaded, or filled with ballast water, the incoming cargo or water displaces hydrocarbon vapor from the compartment into the atmosphere.

6.2.7.1 Loading Loss

When ships and barges are loaded, the incoming cargo displaces hydrocarbon vapor from the compartments into the atmosphere. Two distinct sources contribute to the total loading loss. The emissions during the early stages of loading are composed primarily of the vapor that was present in the tank prior to loading. This vapor originates from previous cargo evaporation. This is referred to as the arrival component. In addition, evaporation of hydrocarbon vapor occurs during loading. This is called the generated component. As a result of evaporation during vessel loading, the total volume of vapor emitted (at compartment temperature and pressure) is typically greater than the volume of liquid loaded. This additional growth is referred to as vapor growth.



--`,,-`-`,,`,,`,`,,`---

8 CHAPTER 19.4

6.2.7.2 Ballasting Loss

Hydrocarbon vapors are displaced into the atmosphere when a compartment that had previously carried a volatile petroleum product is loaded with ballast water. The emissions consist of the vapor present in the ullage space at the start of ballasting together with additional vapors that are generated during the ballasting operation. Note that the ullage space is the space in the compartment above the liquid. The concentration of vapor in the ullage space before ballasting occurs is directly related to the volatility of the previous cargo.

A portion of the hydrocarbons in the compartment will dis- solve into the ballast water. These components will also reach equilibrium in the vapor space of the compartment, and will be displaced as the compartment is filled with ballast water. In addition, a portion of the stock components that are dissolved in the ballast water will be volatilized as the water is removed from the compartments.

7 Description of Methodologies

7.1 INTRODUCTION

This section presents three methods, Level 1, Level 2, and Level 3, for speciating emissions from petroleum operations. Levei i uses representative or typical liquid sample data to calculate the individual constituent concentrations in the vapor phase. Level 1 is the most commonly used method of the three levels; therefore, it is discussed first and in the great- est amount of detail.

The discussions for each level include the methodology and equations used, the advantages, disadvantages, and relative accuracy.

7.2 LEVEL 1 (AVALABLE REPRESENTATWE LIQUID SAMPLE ANALYSIS)

Typically, representative tank vapor profiles are not available, and it is not feasible to collect a vapor sample. In most situations, it is adequate to use a typical liquid profile to calculate the concentrations of constituents in the vapor phase. In some instances, a representative liquid sample must be collected and analyzed. In the latter case, proper sampling procedures and appropriate methods of analysis should be used. Once the liquid component concentrations are known, the constituents in the vapor phase can be approximated by using a vapor-liquid relationship such as Raoult’s Law. More complex approaches should be used when some constituents are not typical hydrocarbons and their liquid and vapor relationships deviate widely from ideal relationships at low concentrations. Examples of compounds that have non-ideal interactions with hydrocarbons include polars such as water, alcohols, and similar oxygenates.

7.2.1 Raoult’s Law

In a multicomponent gas-liquid system at equilibrium, the compositions of the two phases at a given temperature and pressure are directly related. The composition of the vapor phase is uniquely determined by the physical properties of the components in the liquid phase [6,1 i].

Raoult’s Law and Dalton’s Law can be used to estimate the composition of the vapor in equilibrium above a liquid where constituents form a near ideal mixture in both the liquid and vapor phases [ 121.

Raoult’s Law states:

P; = Pi” x; (1)

Where:

Pi = partial pressure of component i in the vapor phase

Pi” = vapor pressure of component i at the equilibrium

xi = molar composition of component i in the liquid

(psia).

temperature (psia).

(mole fraction).

Dalton’s Law extends this relationship to the total compo-

Partial pressure, Pi, is defined as: sition of the vapor phase.

Pi = yip

Where:

Pi = partial pressure of component i in the vapor phase

yi =molar composition of component i in the vapor

P = total pressure of the system analyzed (psia).

(psia).

phase (mole fraction).

Combining Raoult’s Law and Dalton’s Law yields an expression which can be used to predict the vapor phase com- pos¡ tion:

y; = Pi” x/P (3)

where the terms are as previously defined above.

Raoult’s Law is an approximation that is generally valid when the liquid components are chemically similar, such as paraffinic hydrocarbons, and when the total pressure of the system is atmospheric or less [6].

In systems where Raoult’s Law does not apply, the phase composition can be predicted using component vapor-liquid equilibrium constants, Ki, defined by the relationship [IO]:

yi = Ki xi (4)



--`,,-`-`,,`,,`,`,,`---


Where:

Ki = vapor-liquid equilibrium constant of component i (dimensionless).

Ki is a function of the system temperature, pressure, and composition. When Raoult's Law applies, Le., for an ideal liquid, Ki equals Pi" /P.

7.2.2 Application of Raoult's Law

Studies performed for the API (summarized in Appendix A) show that when the liquid composition is known, Raoult's Law will adequately characterize the vapor phase composition for multicomponent hydrocarbon mixtures such as crude oils and gasoline [ 141. Raoult's Law can be used when a representative liquid composition is available by performing the following steps:

Step I . Determine the total VOC emissions using the methods described in API Publications 2514A [i], 19.1 [3], or 19.2 [24], as shown in Figure 1-1.

Step 2. Determine the liquid composition of each stock. As mentioned earlier, the liquid compositions are determined either by using a representative profile or by analysis(es) of a representative stock liquid sample. Table 3 presents an illustrative stock liquid profile for a U.S. mid-continent crude oil.

Step 3. Calculate the mole fraction, xi, of each component in the liquid phase. The concentrations of liquid constituents are generally reported in terms of volume or mass concentrations [vi (volume fraction), or wi (mass or weight fraction)]. It will be necessary to convert from a volume or mass basis to a mole basis to determine the mole fractions. Equation B-8 in Appendix B shows the conversion from a mass concentration, where the molar concentration of a component in the liquid equals its liquid mass concentration multiplied by the molec-

ular weight of the total liquid, divided by the molecular weight of the component [i.e., xi = wi (M, / M i ) ] . The conversion from a mass basis is illustrated in Table 5. Conversions from a volume basis were shown in the example in Section 4.

Step 4. Calculate the vapor pressure of each pure component, P,". Vapor pressures can be calculated using the Antoine Equation at the stock liquid temperature, T, as shown in Appendix B, Equation B-3, where the log of the vapor pressure is a function of the three Antoine Constants, i.e., Log,, Pi = A-[B/(T+C)]. Table 6 lists the Antoine Constants, the calculated vapor pressure using these Antoine Constants, and the vapor pressure found from tables for each of the components in the illustrative crude oil stock. The temperature is assumed to be 60°F. The non-volatiles are arbitrarily assumed to be equal parts of eicosane and biphenyl. Note that the Antoine Constants listed in Table 6 are for use with Equation B-3 where the temperature of the liquid, T, is in degrees Cel- sius and the vapor pressure of component i, Pi, is in millime- ters of mercury.

Notice that in Table 6 the vapor pressure calculated from the Antoine Constants does not differ significantly from the published [ 181 vapor pressure.

Table B-3 in Appendix B lists Antoine Constants for many compounds of interest in petroleum operations. As an alternative, the vapor pressures of a number of hydrocarbons can be read directly from Table B-2 in Appendix B.

Step 5. Determine the partial pressure, Pi, and mole fraction of each component in the vapor phase, y;. According to Raoult's Law the equilibrium partial pressure of a component in the vapor phase, Pi, at a given temperature is its vapor pressure, P,", at that temperature multiplied by its mole fraction in the liquid phase, xi. The mole fraction in the vapor phase is the partial pressure, Pi, divided by the total system pressure, P. To calculate the composition of only the hydrocarbon in

Table 5-Conversion of Liquid Concentrations to Liquid Mole Fractions

Concentration w, M, x, Liquid

(mdrnl) (weight fraction) (lb/lb-mole) (mole fraction) Butane 5900 0.0070 58.12 0.05236 n-Hexane 9060 0.0107 86.17 0.05423 Benzene 2560 0.0030 78.11 0.01691 Cyclohexane 7700 0.009 I 84.16 0.047 19 Toluene 12200 0.0145 92.13 0.0683 1 Octane 10500 0.0124 114.22 0.04742 Ethylbenzene 6420 0.0076 106.16 0.03119 m, p-Xylenes 9020 0.0 107 106.16 0.04383 o-Xylenes 11500 0.0136 106.16 0.05588 Isopropylbenzene 21 10 0.0025 120.19 , 0.00906 I , 2,4-Tnmethlbenzene 12000 0.0142 120.20 0.05150 Napthalene 2740 0.0032 128.16 0.01 103 Non-volatiles 75 1590 0.8915 756.89 0.5111

Total 843300 1.OOO 435.00 1.oooO

Source: Adapted from Reference 14.



--`,,-`-`,,`,,`,`,,`---

10 CHAPTER 19.4

the vapor phase, use the total hydrocarbon vapor partial pressure as the system pressure. Typically, this is the true-vapor pressure of the liquid. To calculate the composition of the entire vapor space, including the air, use the atmospheric pressure as the system pressure. Combining these relationships yields:

y;= P," x;/P

Where:

y; =molar concentration of component i in the vapor (mole fraction).

Pio = vapor pressure of component i at the equilibrium temperature (psia).

xi =molar concentration of component i in the liquid (mole fraction).

P,," = total hydrocarbon vapor partial pressure (liquid true vapor pressure), or atmospheric pressure (psia).

Step 6: Determine the weight fraction of each component in the vapor phase, 2,. Multiply the mole fraction of each component by its molecular weight, divided by the average molecular weight of the vapor.

.~

Where:

zi = weight concentration of component i in the vapor (weight fraction).


Mi = molecular weight of component i (1bAb-mole).

M , = hydrocarbon vapor molecular weight when calculating the concentration of the hydrocarbon vapor phase only, or the average molecular weight of the entire vapor space, including air, when estimating the concentration of the entire vapor space (1bAb- mole), i.e., the sum of each yi times Mi.

The vapor molecular weight, M , is usually unknown, but can be calculated for the hydrocarbon vapor, or the entire vapor space, by using Equation B-6 in Appendix B. The molecular weight of the entire vapor space is the average of the hydrocarbon vapor plus air, which usually averages close to the molecular weight of air of 29 1bAb-mole. For this illustration the hydrocarbon plus air M, is calculated to be 32.6 lb/ lb-mole.

Steps 5 and 6 are illustrated in Table 7 for estimating the concentration of the entire vapor phase.

Step 7. Calculate the total weight fraction of all hydrocarbons in the vapor, z;. Sum the weight fractions of all individual hydrocarbons.

2,=XZ,

Where:

z, = total weight concentration of hydrocarbons in the vapor (weight fraction).

Table &Mid-Continent Crude Oil Vapor Phase Speciation Example

Antoine Vapor Pressure Constants" Range for Anto- Vapor Pressure Vapor Pressure ine's Constants at 60°F at 60°F

Species A B C ("CY (psiay (psia)d Butane 6.80896 935.86 238.73 (-)77-19 25.96 26.28* Hexane Benzene Cyclohexane Toluene Octane Ethylbenzene m,p-X ylenesb o-Xylene Isopropylbenzene I , 2.4-Trimethylbenzene Naphthalene Eicosane Biphenyl

6.87601 6.90565 6.841 30 6.95464 6.91868 6.957 19 6.99980 6.99891 6.93666 7.04383 7.01065 7. i522 7.24541

1171.17 121 i .O33 1201.53 1344.80 1351.99 1424.255 1 457.848 1464.679 1460.793 1573.267 1733.71 2032.7 1998.725

224.41 220.79 222.65 219.48 209.15 213.21 215.21 2 13.69 207.78 208.56 201.86 132.1 202.733

(-)25-92 8-103 20-8 I 6-137

26-164

32-172

19-152

27- 1 66

39-181 52-198 86-250 198-379 69-27 1

1.91 1.17 1.21 0.33 0.15 0.10 0.093 0.0785 0.048 0.020 0.002

< O . O I C

<o.o 1

i .876 1.16 1.218 0.309 0.15* o. IO* 0.093* 0.072* 0.047* 0.022*

<0.02* <0.02' <0.02*

a. Antoine Constants from Reference 5. b. Antoine Constants listed are the average of the Antoine Constants for m-xylene and p-xylene. c. Calculated vapor pressure via the Antoine Constants listed. d. Vapor pressure from Table B-2 in Appendix B or from Reference 18. Those from Reference I8 are designated by an *. e. Vapor pressures via the Antoine Constants listed yield a negative value. Since this is not physically possible, a vapor pressure of <0.01 psia was assigned. f. Eicosane was not listed in Reference 18; however, from the vapor pressures of similar compounds, it was determined that the vapor pressure of eicosane is

<0.02 psia.



--`,,-`-`,,`,,`,`,,`---


Table 7-Calculation of Partial Pressures, Vapor Mole Fractions, and Vapor Weight Fractions

M, 4 pi Y, Z, (IbAb-mole) (mole fraction) at 60°F (psia) (mole fraction) (weight fraction) Species

Butane 58.12 0.05236 1.3594 0.09247 O. 16486 Hexane 86. I7 0.05423 O. i036 0.00705 0.01 864 Benzene 78.11 0.0169 1 0.0198 0.00134 0.0032 1

Toluene 92.13 0.0683 I 0.0225 0.00153 0.00433 Octane 1 14.22 0.04742 0.007 I 0.00048 0.00169 Ethylbenzene 106.16 0.031 19 0.003 1 0.0002 I 0.00069

Cyclohexane 84.16 0.047 19 0.057 1 0.00388 0.01002

m,p-Xylenes 106.16 0.04383 0.0041 0.00028 o.Ooo91 o-Xylene 106.16 0.05588 0.0044 0.00030 0.00009 Isopropylbenzene 120.19 0.00906 O.OOO4 O.ooOo3 o.oO011 1,2,4-Tnmethylbenzene 120.20 0.05150 0.00 I o O.ooOo7 0.00026 Naphthalene 128.16 0.01 103 o.ooo02 0.000001 0.000005 Non-volatiles 756.89 0.51 1 I O.ooo0 0.00000 0.00000

Total liquid 435 1.OOO Total vapor+air 32.6

zi = weight concentration of component i in the vapor (mass fraction).

In this illustration for the total vapor space (hydrocarbon

z, = 0.20479 weight fraction.

plus air):

To calculate the mole fraction concentration y; of only the hydrocarbon portion of the vapor phase, each of the component partial pressures have to be divided by the total hydrocarbon partial pressure of 1 S826 pisa. Weight fractions have to be calculated by using the average molecular weight of only the hydrocarbon vapor from Equation 5 and the hydrocarbon portion mole fractions.

Step 8. Calculate the speciated emissions. This is calculated by multiplying the total hydrocarbon emissions, E, from Step 1 by the mass fraction of qach species of hydrocarbon in the vapor.

Ei = E (z;) (8)

Where:

E; = mass emissions of component i (lb/yr).

E = total hydrocarbon mass emissions (lb/yr).

zj = weight concentration of component i in the hydro- I carbon portion of the vapor (mass fraction).

If the vapor concentration calculations are made only for the hydrocarbon portion of the vapor (using the system pressure as the total hydrocarbon partial pressure), then zi comes directly from that calculation.

If the vapor concentrations were calculated for the entire vapor space, including air (using atmospheric pressure as the system pressure), as per the example in Table 7, then zj for

1.5826 0.10764 0.20479

Equation 7 would be the zi from Table 7, divided by the sum of the z; (z,) from Table 5, or E; = E (z;/z,).

Examples for speciating emissions using these eight steps are included in Section 8.

7.2.3 Precision, Accuracy, and the Variability of Methods

Raoult’s Law was examined for precision and accuracy in predicting vapor phase composition for a number of petroleum feedstocks and gasolines. Details of the study are given in Appendix A and summarized below.

7.2.3.1 Gasolines

Winter and summer blend gasolines were studied [13]. Raoult’s Law predicted the vapor phase concentrations within 79 percent of the measured values; generally underpredicting the winter blend vapors by 63 percent and overpredicting the summer blend vapors by 36 percent. The Peng-Robinson (PR) and Redlich-Kwong-Soave (RKS) equations of state were found to be comparable to Raoult’s Law in their accuracy. There were no significant differences between the vapor phase compositio s predicted by Raoult’s Law and those predicted by either t e PR or RKS equations of state.

7.2.3.2 Heavier Stocks

1

d When heavier petroleum stocks, such as crude oils, jet fuel,

and fuel oil were studied, the vapor phase compositions that were estimated using Raoult’s Law were found to be approxi- mately 5 percent below those predicted by the RKS equation of state, and about 18 percent below those predicted by the PR equation of state [14]. However, for many of the species measured, the vapor concentrations that were predicted by Raoult’s Law appeared to be within the confidence intervals



--`,,-`-`,,`,,`,`,,`---

12 CHAPTER 19.4

computed from the precision and accuracy of the analytical techniques that were used in the project.

The liquid molecular weight was the most important factor affecting the predicted vapor phase compositions. This value varied by up to a factor of 5 with different determination methods, and directly affected the calculated vapor concentrations.

7.2.4 Deviations From Raoult?s Law

Vapors from some petroleum stocks often display marked variations from Raoult?s Law. In these cases, it might be necessary to sample and speciate the vapor phase directly, or to experimentally develop and utilize a relationship such as the following:

(9) Where:

y; =molar concentration of component i in the vapor

Ki = vapor-liquid equilibrium constant at equilibrium

(mole fraction).

pressure and temperature (dimensionless).


Some workable guidelines have been proposed for determining deviations from Raoult?s Law. These guidelines are based on the idea that liquid properties are related to the degree of bonding between molecules. The hydrogen bond is the most important bond and serves as the most important cri- terion. This concept gives rise to five classifications of compounds [8]:

Class I : Liquids capable of forming three-dimensional net- works of strong hydrogen bonds. This includes water, glycol, glycerol, amino alcohols, hydroxy acids, amides, and so forth.

Class 2: Other liquids composed of molecules containing both active hydrogen atoms and donor atoms of oxygen, nitrogen, and fluorine. This includes alcohols, acids, primary phenols, oximes, primary and secondary amines, ammonia, hydrogen fluoride, hydrogen cyanide, nitro compounds with alpha-hydrogen atoms, and so forth.

Class 3: Liquids that contain donor atoms but do not contain active hydrogen atoms. This includes ethers, aldehydes, esters, ketones, tertiary amines, and nitro compounds and nitriles without alpha-hydrogen atoms, and so forth.

Class 4: Liquids composed of molecules with active hydrogen atoms, but no donor atoms. These include molecules with two or three chlorine atoms on the same carbon atom as a hydrogen atom, or one chlorine atom on the same carbon and one or more chlorine atoms on adjacent carbon atoms.

Class 5: All other liquids (those without hydrogen-bond forming capability). These include hydrocarbons, carbon dis-

ulfide, mercaptans, halohydrocarbons not in Class 4, and non- metallic elements.

7.2.4.1 Mixtures of Hydrocarbons and Water

Mixtures of Class 1 and Class 5 compounds, such as .a hydrocarbon in water mixture, can be expected to have posi- tive deviations from Raoult?s Law (e.g., more hydrocarbon in the vapor phase than would be predicted from Raoult?s Law). In some cases, as when the Class 5 compounds are soluble in the Class 1 fluid at low levels (e.g., dilute concentrations of hydrocarbons in water), Henry?s Law can be applied. Henry?s Law is depicted by the relationship:

y; = Hi x,/P,

Where:



HI = Henry?s Constant for component i in the liquid at the equilibrium temperature (psia).

Henry?s Law is valid for solutions in which xi is close to zero, e.g., dilute solutions of i, provided component i does not dissociate, ionize, or react in the liquid phase.

Henry?s Constant can often be represented as the vapor pressure of the component corrected by an activity coefficient, representing its affinity for the solvent. When the component is mixed ideally within the solvent (e.g., two similar hydrocarbons), the activity coefficient is equal to one, and the equation reverts to Raoult?s Law.

When the concentration of the components exceeds their solubility in the liquid, they may begin to form a second phase in the liquid, e.g., oil floating on top of water. In these cases, the vapor liquid equilibrium will become more complex and the applicability of either Raoult?s Law or Henry?s Law will depend on the size and location of the additional phases. Such tankage situations need to be treated as special cases, in which simplified approaches are not likely to give adequate results.

7.2.4.2 Mixtures of Hydrocarbons and Oxygenates

The vapor phase composition above mixtures of Class 5 and Class 2 compounds, such as alcohols, or Class 3 compounds, such as ethers, will also deviate from estimates developed from Raoult?s Law. In these cases Equation 4 applies, and the equilibrium constants, Ki, for each component should be developed to define the relationship between the liquid and vapor concentrations.

A recent study [17] of a simulated gasoline mixture was performed for the API. Equilibrium constants were compared



--`,,-`-`,,`,,`,`,,`---


for butane and benzene with no oxygenate addition and after addition of methyl tertiary-butyl-ether (MTBE) in one case and after addition of ethanol in another case. In both cases, temperatures ranged from 77-140°F. With addition of MTBE ranging from 10 to 20 percent, the equilibrium constants for butane and benzene varied by less than 10 percent. However, the results were more variable when ethanol was added instead of MTBE. For the addition of ethanol ranging from 8.9 to I O percent, the equilibrium constants increased by less than 10 percent for butane, and varied from a 68 percent increase with 8.9 percent ethanol addition to a 15 percent decrease at 10 percent ethanol addition. Overall, the variation of the equilibrium constants was small, relative to expected variations associated with sample collection and analysis. Therefore, while Equation 4 can be used when samples are collected and analyzed, Raoult’sLaw also provides a reasonable estimate of the vapor composition.

Appendix C presents comparisons of molecular weight, boiling point, vapor pressure, and RVP blending values (essentially TVP at 100’F) for various hydrocarbons and oxygenates in the motor gasoline boiling range. These may be used to estimate tank vapor compositions in the absence of more detailed data. Companies might also want to substitute their own RVP blending values when making such calculations.

7.3 LEVEL 2-VAPOR PROFILE AVAILABLE Many of the problems associated with converting liquid

composition data to vapor concentrations could be avoided if vapor profiles were available for systems handling petroleum stocks. Such profiles might resemble the one presented in Table 8 for a simulated 11 component gasoline at 77°F. These profiles must be developed for each individual location. Too often, however, profiles in the literature provide very limited or no information on some of the important hydrocarbon species present in low concentrations. Frequently, they are not derived from actual measurements; instead they are taken from engineering evaluation of literature data. However, once vapor profiles have been measured or calculated for a system, they should be appropriate for future applications with the same petroleum stock.

7.4 LEVEL +VAPOR SAMPLE COLLECTED The best alternative to speciating hydrocarbon emissions is

to directly sample the vapor emissions from the modeled system. However, there are multiple problems associated with direct vapor phase sampling. It should be recognized that such samples are only snapshots of the system, representing only the conditions under which the samples were collected, and are not necessarily applicable to the average conditions associated with the operation. The logistics of collecting vapor samples from the variety of emission points associated with a fixed or floating-roof tank are difficult and often impossible. The vapors in tanks might be stratified, creating

Table &-Vapor Profile for Simulated Gasoline

y,, Mole Fraction Vapor at 77’F

Species Liquid 4.4 psia 14.7 psia

xi Mole Fraction

Butane 0.018 0.123 0.004 2-Methylbutme 0.202 0.627 0.188

Benzene 0.024 0.0 18 0.005 2,2,4-trimethylpentane 0.166 0.057 0.0 17

Methylcyclohexane 0.073 0.024 0.007 Toluene 0.121 0.029 0.009 p-Xylene 0.183 0.0 I6 0.005 Undecane 0.035 O.OO0 0.OOO

2-Methylpentane 0.052 0.068 0.001

1 -Heptene 0.104 0.038 0.01 I

Hexylknzene 0.022 O.OO0 0.OOO

Mole fraction data source: Reference 17.

difficulties in obtaining a representative sample, and the analysis of vapor samples is generally less accurate than that of liquid samples.

7.5 COMMON MISTAKES There are a number of common mistakes that can arise

when attempting to calculate the composition of individual hydrocarbons in the emissions from a storage tank or a marine vessel transfer operation. Several of these mistakes are discussed in the following paragraphs.

7.5.1 Inadequate Stock Characterization

Many of chemicals of interest in a petroleum stock might be present at low concentrations. Frequently these may be ignored for process engineering purposes, but they can be important when assessing emissions of specific chemicals. It is expected that all species more volatile than naphthalene should be measured down to a concentration level of 0.1 weight percent. Few chemicals less volatile than naphthalene will be emitted from the tank. Quantifying these components in the stock will be of value primarily for estimating the molecular weight, or the weight fraction, of the liquid mixture.

7.5.2 Inadequate Molecular Weight Characterization

The average molecular weight of the petroleum feedstock greatly influences the conversion of the measured concentrations in the liquid phase from a weight basis to a molar basis, and will directly affect the estimate of the vapor phase composition. The average molecular weight of the liquid stock must be determined accurately.

7.5.3 Failure to Calculate the Vapor Phase Composition

Once the liquid is characterized, it might be assumed that the vapors have the same relative composition as the liquid.



--`,,-`-`,,`,,`,`,,`---

14 CHAPTER 19.4

However, this is a faulty assumption due to the variation in volatility between the individual components. Only clingage losses are characterized by the liquid composition. In all other cases, even though the vapor phase contains the same species as the liquid, the relative composition will vary significantly. In addition, some of the less volatile components will essentially be non-detectable in the vapor phase.

7.5.4 Failure to Convert Composition From Weight to Molar Units

The calculation procedure requires that the liquid concentration that is on a weight basis be converted to a molar basis, and that the calculated vapor composition that will initially be on a molar basis be converted back to a weight basis. The need for these conversions is often forgotten, or they are performed incorrectly.

7.5.5 Attempting to Determine the StockTrue- Vapor Pressure From the Composition of the Vapor Phase

While Raoult’s Law can be used to estimate the vapor phase composition, it is much less successful in accurately predicting the true vapor pressure of the mixture. This must be determined separately, usually by direct measurement of representative examples.

7.5.6 Definition of System Pressure Not Consistent With Objective of the Estimate

Before making calculations, it is important to decide whether the results will be used to speciate just the hydrocarbon portion of the vapor (using total hydrocarbon partial pressure as the system pressure), or the entire vapor space including the &r (using atmospheric pressure). Conversion from one type of estimate to the other is fairly simple, but it is more efficient to determine how the results will be used prior to calculation.

8 Example Applications

8.1 EXAMPLE 1: FIXED-ROOFTANK This example is taken from MPMS 19.2.

8.1.1 Problem

Estimate the total annual evaporative loss of hydrocarbons and individual components, in pounds per year, given the following information.

A fixed-roof tank has the following characteristics:

a. A diameter of 100 feet. b. A shell height of 40 feet. c. A cone roof with a roof slope of 0.0625 feet per foot. d. A liquid level of 20 feet.

e. A maximum liquid level of 38 feet. f. The tank is painted white, and the paint is in satisfactory condition. g. The breather vent pressure setting is 0.03 pounds per square inch gauge, and the breather vent vacuum setting is -0.03 pounds per square inch gauge.

The product stored in the tank has the following characteristics:

a. A stock of U.S. mid-continent cmde oil. b. An RVP of 2.0 pounds per square inch. c. The stock vapor and liquid composition are not given. d. An annual net throughput of 1.5 million barrels per year.

The ambient conditions are as follows:

a. A daily maximum ambient temperature of 70°F. b. A daily minimum ambient temperature of 50°F. c. A total daily solar insolation on a horizontal surface of 1200 British thermal units per square foot day. d. An atmospheric pressure of 14.7 pounds psia.

8.1 -2 Solution

Table 9 presents a summary of the required steps for calculating the individual component emissions. The component cyclohexane will be used to illustrate each step that is required in the individual calculations described below:

Step I . Determine the total hydrocarbon emissions by using the methods described in API hblication 19.2 1241. Using API Publication 19.2 on pages 26-28, the estimated hydrocarbon standing storage loss from this tank was found to be 1 1,490 pounds per year, while the hydrocarbon working loss was estimated at 47,820 pounds per year. The total hydrocarbon loss, which is the sum of the standing storage loss and the working loss, was found to be 59,310 pounds per year.

Step 2. Determine the liquid weight concentration, wi, of each component in the stock. The composition of the stock was not given; therefore, a standard liquid stock profile for a U.S. mid-continent crude oil will be used [14]. The liquid weight compositions, in terms of weight fraction, are given in Table 9. The liquid weight fraction of cyclohexane is 0.009 I .

Step 3. Calculate the liquid mole fraction, xi , of each component in the liquid phase. Before the liquid mole fractions can be calculated, the liquid average molecular weight must be determined. Equation B-5 in Appendix B can be used if the liquid composition, w;, of each component is known.

M, = 1 E( Wi/Mi) (1 1)

In this case, the liquid average molecular weight was given as 435 (1bAb-mole).

Equation B-8 in Appendix B is then used to calculate the liquid mole fractions, xi. The liquid mole fraction of cyclohexane is the liquid weight fraction (0.0091) multiplied by



--`,,-`-`,,`,,`,`,,`---


the liquid average molecular weight (439, divided by the molecular weight of cyclohexane (84.16). Therefore, the liquid mole fraction of cyclohexane is 0.047.

Step 4 . Determine the vapor pressure of each component, P,". The vapor pressure of cyclohexane can be calculated from its Antoine Constants, shown in Table 9 and Equation B-3 in Appendix B. The vapor pressure of cyclohexane at 60°F is 1.21 psia.

Step 5. Determine the partial pressure, Pi, and vapor mole fraction, yi, of each component in the vapor phase. The partial pressure of a component in the vapor phase, Pi, is its vapor pressure, Pio, multiplied by its mole fraction in the liquid phase, xi. This is illustrated as Raoult's Law in Equation 1. The vapor mole fraction, y;, is the partial pressure, Pi, divided by the atmospheric pressure, Po.

The partial pressure, Pi, of cyclohexane is Pio (1.21 psia) multiplied by xi (0.047), which is 0.057 psia.

The vapor mole fraction, y;, of cyclohexane in the total vapor space (including air) is Pi (0.057 psia) divided by atmospheric pressure, P, (14.7 psia), which equals 0.0039. The vapor mole fraction of cyclohexane in only the hydrocarbon portion of the total vapor is Pi (0.057 psia, rounded to 0.06 in Table 9), divided by the total partial pressure of the hydrocarbons, C Pi (1.59 psia), which equals 0.036.

Step 6. Determine the weight fraction, zj of each component in the vapor phase. The vapor weight fraction, z;, is the vapor mole fraction, y;, multiplied by its molecular weight, Mi, divided by the molecular weight of the vapor, M,.

The average molecular weight of the vapor, M,, can be calculated from Equation 12.

M, = c(y; M;)

The average molecular weight of the total vapor (hydrocarbon plus air) is 32.6. By leaving out the term for the air portion, the same equation could be used to find the average molecular weight of just the hydrocarbon portion of the vapor, which equals a molecular weight of 62.4.

The weight fraction of cyclohexane in the total vapor phase is (0.0039)(84.16)/(32.6) = 0.01. The weight fraction in the hydrocarbon portion of the vapor is (0.036)(84.16)/(62.4) = 0.0486.

Step 7. Calculate the total weight fraction, z,, of all hydrocarbons in the total vapor space. This is the sum of all the hydrocarbon weight fractions. In this example the sum is 0.206 1.

Step 8. Calculate the individual speciated emissions. Multi- ply the total hydrocarbon emissions, E, from Step 1 by the weight fraction of each species, z;, and divide by the total weight fraction of the hydrocarbons in the vapor phase, z,.

The total hydrocarbon emissions from Step 1 is 59,310 pounds per year. The weight fraction of cyclohexane from Step 6 is 0.01, and the total weight fraction from Step 7 is 0.2061. Therefore, the cyclohexane emission is 2,882 pounds per year, which represents about 5 percent of the total hydrocarbon emissions. The same answer is obtained by multiplying 59,3 1 O by the fraction of cyclohexane in the hydrocarbon portion of the vapor or (59310)(0.0486) = 2882.

Table %Fixed Roof Tank Speciated Emissions for a U.S. Mid-Continent Crude Oil

Hexane Benzene Cyclohexane Toluene Octane Ethylbenzene m,p-Xylenes o-Xylene Isopropy Ibenzene

0.0107 0.0030 0.009 1 0.0145 0.0124 0.0076 0.0107 0.0136 0.0025

0.0142 1,2,4- Trimethylbenzene Naphthalene 0.0032 Non-volatiles 0.8915

Total 1.0

"Source: Reference 14. bSource: Reference 5. 'Temperature of system is 60°F.

86.17 78.1 1 84.16 92.13

114.22 106.16 106.16 106.16 120.19

120.20

128.16 756.89 435

0.054 0.017 0.047 0.068 0.047 0.03 1 0.044 0.056 0.009

0.05 1

0.01 1 0.512 1 .o

6.87601 6.90565 6.84130 6.95464 6.91968 6.95719 6.99980 6.99891 6.93666

7.04383

7.01065 - -

117.17 12 1 1 .O33 1201.53 1344.80 1351.99 1424.255 1457.848 1464.679 1460.793

1573.267

1733.71 - -

224.41 220.790 222.65 219.48 209.15 213.21 215.21 213.69 207.78

208.56

201 3 6 - -

Antoine Constantsh p,o p , w, 4 P," in mm Hg, Tin "C Vapor Partial Y, r, E,

(weight M, (mole Pressure' Pressure (mole (weight Emissions Species fraction)' (Ib/lb-mole) fraction) A B C (psia) (psia) fraction) fraction) (Ibdyr)

Butane 0.0070 58.12 0.052 6.80896 935.86 238.73 26.01 1.36 0.093 0.1653 47,554 1.91 0.10 0.007 0.0186 5349 1.17 0.02 0.0013 0.0032 917 1.21 0.06 0.0039 0.0100 2882 0.33 I 0.02 0.0015 0.0044 1253 O. 154 0.01 0.0005 0.0017 500 0.104 <0.01 0.0002 0.0007 207 0.093 <0.01 0.0003 0.0009 260 0.0788 <0.01 0.0003 0.0010 280 0.0481 <0.01 O.ooOo29 0.00011 31

0.0204 <0.01 0.000071 0.00026 76

0.0021 <0.01 0.000001 0.000006 2 - - - - - - 1.59 0.1079 0.2061 59,310



--`,,-`-`,,`,,`,`,,`---

16 CHAPTER 19.4

8.2 EXAMPLE 2: INTERNAL FLOATING-ROOF TANK

This example is taken from Section 7.3 of API MPMS 19.2.

8.2.1 Problem

Estimate, in pounds per year, the total annual evaporative loss of hydrocarbon and individual components given the following information.

A freely vented, internal floating-roof tank has the following characteristics:

a. b.

d. e. f.

C.

A diameter of 100 feet. Typical deck fittings. Built-up columns. A floating deck with bolted seams. A column-supported, fixed roof in satisfactory condition. A shell painted an aluminum color.

The product stored in the tank has the following characteristics:

a. A stock of motor gasoline. b. An RVP of 10 pounds per square inch. c. 6.1 poundgallon average liquid stock density (the stock liquid cornposition is given in Table 6). d. 1.5 million bbl/yr average annual net throughput.


a. 60°F average annual ambient temperature. b. 14.7 psia atmospheric pressure.

8.2.2 Solution

A summary of the steps required in calculating the individual component emissions is shown in Table 10. The speciated composition of the standing storage loss is based on the composition of the vapor in equilibrium with the stock liquid. The speciated composition of the withdrawal loss is based on the composition of the stock liquid itself. The component 2- methylbutane is used to illustrate each required step in the following individual calculations.

Step I . Determine the total hydrocarbon emissions using the methods described in API Publication 25 19 [4]. From API Publication 25 19, page 6, the standing storage loss consisting of the rim seal loss, the total deck fitting loss, and the deck seam loss was estimated at 15,490 pounds per year, while the withdrawal loss was estimated at 138 pounds per year.

Step 2. Determine the liquid weight concentration, wi, of each component. The liquid weight concentrations, wi, in terms of weight fraction, are presented in Table 8. The weight fraction of 2-methylbutane is 0.202.

Step 3. Calculate the mole fraction, xi, of each component in the liquid phase. Before the mole fractions can be calculated, the average molecular weight of the liquid, M,, must be determined by using Equation 8.

The average molecular weight is 93.7. Equation B-8 in Appendix B is used to calculate the liquid

mole fractions, xi. The mole fraction of 2-methylbutane is the weight fraction (0.202) multiplied by the average molecular weight (93.7), divided by the molecular weight of 2-methylbutane (72.15). Therefore, the mole fraction of 2-methylbutane is 0.262.

Table 1 O-Internal Floating-Roof Tank Speciated Emissions for a Simulated Gasoline

Antoine Constantsh E, w, M, x, P," in mmHg, Tin "C p, O p , Y , Z, E, Emissions

(weight (IbAb- (mole Vapor Partial (mole (weight Emissions Withdrawal Species fnctionp mole) fraction) A B c Pressurec Pressure fraction) fraction) (Ibdyr) (Ibdyr)

Butane 0.018 58.13 0.029 6.80896 935.86 238.73 25.9603 0.75 0.0512 0.0751 2585 2.484 2-Methylbutane 2-Methylpentane Benzene

Trimethyl pentane

Methy

Toluene p-Xylene Undecane Hexylbenzene

Total

2,2,4-

I -Heptane

cylophexane

0.202 0.052 0.024

0.166

0.104

0.073

0.121 0.183 0.035 0.022 1 .o

72.15 86.18 78.12

114.23

98.19

98.19

92.15 106.17 156.32 162.28 93.7

0.262 0.057 0.029

0.136

0.099

0.070

0.123 0.162 0.02 1 0.013 1.0

6.83315 6.839 1 6.90565

6.81189

6.901 87

6.823

6.95464 6.99052 6.9722 - -

1040.73 1135.41 121 1.033

1257.84

1258.345

1270.763

1344.8 1453.43 1569.57 - -

235.45 226.57 220.79

220.74

219.30

22 I .42

2 19.48 215.31 187.70 - -

9.3863 2.46 2.7239 0.15 1.1667 0.03

0.5943 0.08

0.6748 0.07

0.5570 0.04

0.3330 0.04 0.0955 0.02 0.0034 0.00 O.OoO8d 0.00 - 3.65

O. 1675 0.0105 0.0023

0.0055

0.0046

0.0026

0.0028 0.0010 0.000004 0.000000 0.2480

0.3048 0.0228 0.0045

0.0159

0.01 13

0.0065

0.0064 0.0028 O.oooO19 0.000003 0.4501

10489 784 155

546

388

225

22 I 91 1 O

15490

27.876 7.176 3.312

22.908

14.352

10.074

16.698 25.254

4.83 3.036

138.0

"Source: Reference 17. 5ource: Reference 5. 'Temperature of system is 60°E dSource: Reference 7.



--`,,-`-`,,`,,`,`,,`---


Step 4. Determine the vapor pressure of each component, Pi". The vapor pressure of 2-methylbutane can be calculated from its Antoine Constants, listed in Table 7, and from Equa- tion B-3, shown in Appendix B. The vapor pressure of 2- methylbutane at 60°F is 9.39 psia.

Step 5. Determine the partial pressure, Pi, and mole fraction, y;, of each component in the vapor phase. The partial pressure of a component in the vapor phase, Pi, is its vapor pressure, Pi", multiplied by its mole fraction in the liquid phase, xi. This is illustrated as Raoult's Law in Equation 1. The vapor mole fraction, yi, is the partial pressure, Pi, divided by the total atmospheric pressure, Pa.

The partial pressure, Pi, of 2-methylbutane is Pio (9.39 psia) multiplied by xi (0.262), which is 2.46 psia.

The vapor mole fraction, y;, of 2-methylbutane is Pi (2.46 psia), divided by the atmospheric pressure, Pa (14.7 psia), which equals 0.1675.

Step 6. Determine the weight fraction, zi, of each component in the vapor phase. The vapor weight fraction, zi, is the vapor mole fraction, y;, multiplied by its molecular weight, M, divided by the average molecular weight of the vapor, M,

The average molecular weight of the vapor, M,, is 39.7 and can be calculated using Equation 9.

The weight fraction of 2-methylbutane in the vapor phase is (0.1675)(72.15)/(39.7) = 0.305.

Step 7. Calculate the total weight fraction, z,, of all hydrocarbons in the vapor space. This is the sum of all the hydrocarbon vapor weight fractions, z;, which in this example is equal to 0.450.

Step 8. Calculate the individual speciated emissions.

8.2.3 Standing Storage Loss

This is calculated by multiplying the total standing storage loss from Step 1 by the vapor phase weight fraction of each species, zi, and dividing by the total weight fraction of the hydrocarbons in the vapor phase, z,.

The total hydrocarbon emissions from standing storage loss, from Step 1, is 15,490 pounds per year. The vapor weight fraction of 2-methylbutane from Step 6 is 0.305, and the total vapor weight fraction from Step 7 is 0.450. There- fore, the standing storage loss emissions for 2-methylbutane are 10,489 pounds per year.

8.2.4 Withdrawal Losses

This is calculated by multiplying the total withdrawal loss from Step 1 by the liquid phase weight fraction of each species, xi.

The total hydrocarbon emissions from withdrawal loss, from Step 1, is 138 pounds per year. The liquid weight fraction of 2-methylbutane, from Step 2, is 0.202. Therefore, the withdrawal loss emissions for 2-methylbutane are 27.9 pounds per year.

2-methylbutane represents about two-thirds of the standing storage loss, but only 20 percent of the withdrawal loss.

8.3 EXAMPLE 3: EXTERNAL FLOATING-ROOF TANK

This example is taken from Section 7.2 of API MPMS 19.2.

8.3.1 Problem


A welded, external floating-roof tank in satisfactory condition that has the following characteristics:

a. A diameter of 100 feet. b. A shell painted an aluminum color. c. A pontoon floating roof. d. A mechanical-shoe primary seal. e. Typical roof-fittings.

The motor gasoline stored in the tank has the following characteristics:

a. An RVP of 10 pounds per square inch. b. A liquid stock density of 6.1 pounds per gallon. c. An average net throughput of 1.5 million barrels per year.


a. An average annual ambient temperature of 60°F. b. An atmospheric pressure of 14.7 psia. c. An average annual windspeed of 10 miles per hour.

8.3.2 Solution

The speciated composition of the standing storage loss is based on the composition of the vapor in equilibrium with the liquid stock. The speciated composition of the withdrawal loss is based on the composition of the liquid stock itself. Using the representative liquid data included in the prior example, speciated hydrocarbon emissions were again calculated by using the liquid gasoline stock composition for the withdrawal loss, and using Raoult's Law to calculate the vapor composition of the standing storage loss. The results are shown in Table 1 1 . The component 2-methylbutane will be used to illustrate each step required in the individual calculations below.

Step I . Determine the total hydrocarbon emissions using the methods described in API Publication 2517 121. From page 16, the standing storage loss, consisting of the rim seal loss and the total roof fitting loss was estimated at 33,400 pounds per year, while the withdrawal loss was estimated at 129 pounds pcr year.

Step 2. Determine the liquid concentration of each component. The liquid concentrations are presented in Table 9 in



--`,,-`-`,,`,,`,`,,`---

18 CHAPTER 19.4

terms of weight fraction, w,. The weight fraction of 2-methylbutane is 0.202.

Step 3. Calculate the mole fraction, x,, of each component in the liquid phase. Before the mole fractions can be calculated, the average liquid molecular weight, M,, must be determined by using Equation 1.

The average liquid molecular weight is 93.7. Equation B-8 in Appendix B is then used to calculate the

liquid mole fractions, x,. The mole fraction of 2-methylbutane is the weight fraction (0.202) multiplied by the average liquid molecular weight (93.7), divided by the molecular weight of 2-methylbutane (72.15). Therefore, the liquid mole fraction of 2-methylbutane is 0.262.

Step 4. Determine the vapor pressure of each component, Pl". Note that the first three steps are identical to the first three steps in Example 2. The temperature of this system is the same as the temperature of the system in Example 2. There- fore, the vapor pressure of each component will be the same as the vapor pressures found in Example 2. The vapor pressure of 2-methylbutane can be calculated from its Antoine Constants, shown in Table 10, and by using Equation B-3 in Appendix B. The vapor pressure of 2-methylbutane at 60°F is 9.39 psia.

Step 5. Determine the partial pressure, Pl, and mole fraction, y, , of each component in the vapor phase. The partial pressure, P,, of a component in the vapor phase is its vapor pressure, f,", multiplied by its mole fraction in the liquid phase, x,. This is illustrated as Raoult's Law in Equation 1. The vapor mole fraction, y,, is the partial pressure, i',, divided by the total atmospheric pressure, Pu.

The partial pressure, P,, of 2-methylbutane is P," (9.39 psia) multiplied by x, (0.262), which is 2.46 psia.

The vapor mole fraction, yi, of 2-methylbutane is fi (2.46 psia) divided by atmospheric pressure Pu (14.7 psia), which equals O. 1675.

Step 6. Determine the weight fraction, zi, of each component in the vapor phase. The vapor weight fraction is the vapor mole fraction, yi, multiplied by its molecular weight, Mi, divided by the average molecular weight, M,, of the vapor.

The average molecular weight of the vapor, M,, is 39.7 and can be calculated by using Equation 9.

The weight fraction of 2-methylbutane in the vapor phase is (O. 1675)(72.15)/(39.7) = 0.305.

Step 7. Calculate the total weight fraction, z,, of all hydrocarbons in the vapor space. This is the sum of all the vapor weight fractions, zi, which is equal to 0.4501.

Step 8. Calculate the individual speciated emissions.

8.3.3 Standing Storage Loss

This is calculated by multiplying the total standing storage loss from Step 1 by the vapor phase weight fraction of each species, zi, divided by the total weight fraction of the hydrocarbons in the vapor phase, z,.

The total hydrocarbon emissions from standing storage loss, from Step 1, is 33,400 pounds per year. The weight fraction of 2-methylbutane, from Step 6, is 0.305. The total weight fraction from Step 7 is 0.4501. Therefore, the standing storage loss emissions for 2-methylbutane are 22,617 pounds per year.

8.3.4 Withdrawal Loss

This is calculated by multiplying the total withdrawal loss, from Step 1, by the liquid phase weight fraction of each species, x;.

Table 1 1-External Floating-Roof Tank Speciated Emissions for a Simulated Gasoline

E, E, w, p," p," Z, Emissions Emissions

(weight M, & Vapor Partial Y, (weight Evap Withdrawal Species fraction)" (IbAb-mole) (mole fraction) Pressureh Pressure (mole fraction) fraction) (Ibdyr) (Ibs/yr)

Butane 0.018 58.13 0.029 25.9603 0.75 0.05 12 0.0751 5574 2.322 2-Methylbutane 0.202 72.15 0.262 9.3863 2.46 O. I675 0.3048 226 I7 26.058 2-Methylpenthane 0.052 86.18 0.057 2.7239 0.15 0.0 I 05 0.0228 I690 6.708 Benzene 0.024 78.12 0.029 1.1667 0.03 0.0023 0.0045 334 3.096

O. 166 114.23 0.136 0.5943 0.08 0.0055 0.0 I59 I177 21.414 2 ,2 ,4 - Tnmethylpentane 1 -Heptane 0.104 98.19 0.099 0.6748 0.07 0.0046 0.01 13 837 13.416 Methyl cylophexane 0.073 98.19 0.070 0.5570 0.04 0.0026 0.0065 485 9.417 Toluene 0.121 92.15 0.123 0.3330 0.04 0.0028 0.0064 476 15.609 p-Xylene O. I83 106.17 O. I62 0.0955 0.02 0.0010 0.0028 208 23.607 Undecane 0.035 156.32 0.02 1 0.0034 0.00 0.000004 O.ooOo19 I 4.5 I5 Hexylbenzene 0.022 162.28 0.013 O.oo08' 0.00 0.000000 0.000003 O 2.838

Total 1.0 93.7 1 .o - 3.65 0.2480 0.4501 33400 129.0

"Source: Reference 17. hTempenture of system is 60°F. 'Source: Reference 7.



--`,,-`-`,,`,,`,`,,`---


The total hydrocarbon emissions from withdrawal loss, from Step 1, is 129 pounds per year. The liquid weight fraction of 2-methylbutane, from Step 2, is 0.202. Therefore, the withdrawal loss emissions for 2-methylbutane are 26.1 pounds per year.

The 2-methylbutane represents about two-thirds of the standing storage loss, but only 20 percent of the withdrawal loss.

8.4 EXAMPLE 4: MARINE VESSELTRANSFER OPERATIONS

This example is from Section 1.3 of API Publication 2514A [i].

8.4.1 Problem


Vessel description: 30,000 dead-weight ton tanker loading 125,000 barrels of motor gasoline; all compartments receiv- ing gasoline previously carried volatile cargo.

Compartment arrival conditions: 25 percent uncleaned (Category i), 10 percent ballasted (Category 2), and 65 percent gas-freed (Category 4).

8.4.2 Solution

The speciated composition of the total hydrocarbon emissions is based on the composition of the vapor presented in Table 10. The component 2-methylbutane will be used to illustrate each step required in the individual calculations below.

Step 1. Determine the total hydrocarbon emissions using the methods described in API Publication 2514A [i]. Using the methodology described in API Publication 2514A, page 8, total emissions from gasoline loading operations are 6800 pounds per year based on the compartment arrival conditions.

3453 pounds per year are due to the compartment being 25 percent uncleaned, 903 pounds per year are due to the compartment being 10 percent ballasted, and 2444 pounds per year are due to the compartment being 65 percent gas-freed.

Step 2. Calculate the total weight fraction, z,, of all hydrocarbons in the vapor space. This is the sum of all the vapor weight fractions, which is 0.4501.

Step 3. Calculate the individual speciated emissions. Multi- ply the total hydrocarbon emissions calculated from Step 1 by the vapor phase weight fraction of each species, zi. Divide by the total weight fraction of the hydrocarbons in the vapor phase, z,.

The total hydrocarbon emissions from Step 1 are 6800 pounds per year. The weight fraction of 2-methylbutane from Table 12 is 0.305. The total weight fraction from Step 2 is 0.450. Therefore, the speciated emissions for 2-methylbutane are 4605 pounds per year.

Table 12-Marine Vessel Transfer Operations Speciated Emissions for a Simulated Gasoline

Z, (Weight E, Mass Emissions Species Fraction Vapory (Ibdyr)

Butane 0.075 I 1135 2-Methylbutane 0.3048 4605 2-Methylpentme 0.0228 344 Benzene 0.0045 68 2,2,4-Tnmethylpentane 0.0159 240 I-Heptane 0.01 13 i 70 Methyl cylophexane 0.0065 99 Toluene 0.0064 97 p-Xylene 0.0028 42 Undecane O.ooo019 O Hexylbenzene 0.000003 O

Total 0.4501 6800

System temperature is 60°F.



--`,,-`-`,,`,,`,`,,`---

APPENDIX A-VALIDITY OF RAOULT’S LAW

A.l Raoult’s Law Versus Other Equations of State

The validity of using Raoult’s Law, versus using other equations of state, to predict speciated emissions from petroleum storage tanks, was investigated by Radian Corporation for API in 1990 [13]. This study used data for two gasolines, summer blend and winter blend. Vapor compositions were predicted from liquid composition data using Raoult’s Law, the RKS equation of state [ 191, and the PR equation of state [20]. These data were then compared to the measured vapor composition data. Summaries of these results are presented in Tables A-1 and A-2 for summer blend and winter blend gasolines. The measured liquid and vapor concentrations were taken from Wright, Menon, and Peoples [16]; the predicted vapor concentrations were calculated using the ASPEN/SP [21] simulator’s Two-Phase Hash model. Calculations were based on a system temperature of 60°F.

As presented in Table A-1 for summer blend gasoline, there were no significant differences between the vapor compositions predicted by Raoult’s Law and the two equations of state. All three of the methods predicted vapor concentrations that were higher than the measured vapor concentrations. Results presented in Table A-2 for the winter blend gasoline show there is little difference between Raoult’s Law and the equations of state predictions for vapor compositions. How- ever, all three methods predicted vapor concentrations that were lower than the measured values.

From this study, Radian Corporation concluded that the use of Raoult’s Law in the prediction of vapor compositions from liquid composition data provides essentially the same level of accuracy as those based on the two equations of state,

Table A-1-Summary of Results for Summer Blend Unleaded Gasoline

Table A-2-Summary of Results for Winter Blend Unleaded Gasoline

Measured” Predicted VaDor Concentration (wt%) Concentration (wt%)

Toxic Species Liquid Vapor Raoulth PRC RKSd Cyclohexane 0.50 0.34 0.16 0.17 0.16 Benzene 1.82 1.43 0.56 0.71 0.67 Toluene 9.11 2.10 0.82 1.00 0.92 O-Xylene 3.59 0.51 0.00 0.00 0.00 Isomers of xylene 8.75 1.02 0.21 0.23 0.00 Ethylbenzene 2.08 0.31 0.00 0.00 0.00 Isopropylbenzene O. 19 - 0.00 0.00 0.00

4.21 0.05 0.00 0.00 0.00 1,2,4- Trimethylbenzene Napthalene 0.25 - 0.00 0.00 0.00

asOurce: References 16 and 13. hRaoult = Raoult’s Law. ‘PR dRKS

= Peng-Robinson equation of state. = Redlich-Kwong-Soave equation of state.

at least for the gasolines included in the assessment. Because gasoline was the only stock investigated in this study, these results could not be generalized to include other stocks of interest, such as crude oils.

In order to further test the validity of the vapor and liquid equilibrium models investigated in 1990 [ 131, the study was continued for API by Radian Corporation in 1991 and the results were reported in 1992 [ 141. The liquid and vapor equilibrium compositions for selected species in three crude oil samples-mid-continent crude oil, Alaska North Slope ( A N S ) crude oil, and ANS crude oil spiked with natural gas liquids-were analyzed. Two refined product samples, JP-4 Aviation Fuel and No. 2 Fuel Oil were also analyzed. Twelve compounds were selected to be speciated and are listed in Table A-3.

Toxic Species Cyclohexane Benzene Toluene O-Xylene Isomers of xylene Ethylbenzene Isopropylbenzene

Trimethylbenzene I ,2,4-

Measureda Concentration (wt%)

Liquid Vapor 0.58 0.20 1.93 0.73

10.32 0.81 3.39 0.05 9.16 0.21 2.05 0.05 0.19 -

3.52 0.04

Predicted Vapor Concentration (wt%)

Raoulth PRc RKSJ 0.25 0.27 0.26 0.83 1.05 0.99 1.29 1.57 1.45 0.00 0.00 0.00 0.31 0.33 0.30 0.00 0.00 0.00 0.00 0.00 0.00

0.00 0.00 0.00

”Source: References 16 and 13. hRaoult = Raoult’s Law. ‘PR “KS

= Peng-Robinson equation of state. = Redlich-Kwong-Soave equation of state.

Table A-3-Compounds Selected for Speciation

Butane n-Hexane Benzene

Cyclohexane Toluene Octane

Ethylbenzene m, p-Xylenes

o-Xylene Isopropy Ibenzene

1,2,CTrimethylbenzene Naphthalene

21



--`,,-`-`,,`,,`,`,,`---

22 CHAPTER 19.4

Table A-4-GC Analysis Concentrations Using Average Response Factors (ARF) and Linear Regression (LR) for Liquid Phase Samples (Concentrations in pg/mL)

No. 2 Fuel Oil Mid-Cont Crude JP-4 Aviation Fuel (Mean Value) (Mean Value) Alaska Crude Spiked Alaska Crude

Compounds ARF LR ARF LR ARF LR ARF LR ARF LR Butane 4280 4280 e250 e300 5900 5730 2120 2120 5680 5680 n-Hexane Benzene Cyclohexane Toluene Octane Ethylbenzene m,p-Xylenes o-Xylenea Isopropy Ibenzene I~2,4-Trimethylbenzeneb Naphthalene

19900 3880 6030 29500 8330 1590 2530 1730 793 3500 i 820

18400 5010 6200 28100 8510 2740 3600 2380 1820 3800 2080

e250 e250 e250 300 483 I57

1060 477 116 1780 9450

e300 e300 e300 1040 948 793 1620 I 1 0 0 656 1940 9120

9060 2560 7700 12200 10500 6420 9020 11500 21 10 12000 2740

7430 1370 5890 9960 8480 4880 7340 9740 684

10400 i 170

7710 2770 7350 10800 9060 1780 7170 2180 942 9490 1890

6190 I560 5570 8690 7220 490 5560 830

<300 7970 287

8220 2640 6930 9060 7540 I450 5620 1760 812

7630 1530

6650 1440 5190 7100 5840 176

4090 426 e300 6140 e300

'Value might be high due to coelution with nonane. Value might be high due to coelution with decane.

Table A-5-GC Analysis Concentrations Using Average Response Factors (ARF) and Linear Regression (LR) for Vapor Phase Samples (Concentrations in pg/mL)

No. 2 Fuel Oil Mid-Cont Crude JP-4 Aviation Fuel (Mean Value) (Mean Value) Alaska Crude Spiked Alaska Crude

Compounds ARF LR ARF LR ARF LR ARF LR ARF LR Butane 60400 58100 e3 e3 37900 44100 45500 53100 122000 14300 n-Hexane 6800 7970 Benzene 2940 3070 Cyclohexane 1080 1050 Toluene 1110 1430 Octane 175 I45 Ethylbenzene 46.4 e60 m,p-Xylenes 91.1 44 o - X y I e n e a e40 e40 Isopropylbenzene e40 e40 1,2,4-Tnmethylbenzeneb e40 e40 Naphthalene e40 e40

aValue might be high due to coelution with nonane. "Value might be high due to coelution with decane.

14 7

.28 14 14 6 22 5 2 14 2

14 5 27 14 12 2 15 5

e3 5 2

2450 40 i 482 387 204 58 158 39 8 13 21

2930 353 441 409 200 58 i 64 42 4 15 13

3900 962

1450 404 210 52 85.6 21.1 7.5 2.1 9.2

4630 916

1470 383 216 52.1 89.1 21.3 4.7

e3 e3

5460 1530 2740 406 228 57.8 87.9 22.4 7.83 2.38 1.48

6573 1580 2980 378 236 57.6 91.4 22.7

5.1 <3 e 3



--`,,-`-`,,`,,`,`,,`---


The compounds chosen for speciation were selected to reflect compounds that: ( 1 ) are common to the crude oils and products analyzed; (2) represent a wide volatility range; and (3) are listed as Hazardous Air Pollutants (HAPS) in the Clean Air Act Amendments of 1990. Most of the compounds listed in Table A-3 represent air toxics expected to be present in the crude oils and refined products. Butane, hexane, and octane were selected on the premise that they might contribute significantly to the total vapor above the selected crude oils and refined products.

Liquid and vapor phase concentrations of the 12 chosen compounds were determined by gas chromatography/flame ionization detection (GCFID). Tables A-4 and A-5 present the liquid and vapor phase concentrations, respectively, using both average response factors (ARF) and linear regression (LR). Liquid and vapor molecular weights were also measured.

The measured liquid concentrations were used as input to a flash calculation routine in the process simulator ASPENISP [21] to predict the equilibrium vapor concentrations using Raoult’s Law and the RKS and PR equations of state. The results obtained for the vapor concentrations that were predicted by each equation of state were compared to Raoult’s Law. Tables A-6 and A-7 compare the average vapor concentrations that were predicted by the RKS and PR equations of state to those predicted by Raoult’s Law. The tables compare average response factor analytical data and linear regression analytical data. The results are presented in terms of the ratio of the predicted vapor concentration of each equation of state to Raoult’s Law.

The study found that the vapor concentrations predicted by Raoult’s Law and the two equations of state almost agreed.

The RKS equation of state predicted vapor concentrations that were about 5 percent higher than those predicted by Raoult’s Law, and the PR equation of state predictions were about 18 percent higher than Raoult’s Law. The differences within the precision and accuracy of the analytical techniques that were used to obtain the compositions in the liquid phase.

It was concluded that there were no advantages in using the equations of state instead of Raoult’s Law, especially since the equations of state are much more difficult to use than Raoult’s Law. It was observed that “more precise and accurate measurements are possible for liquid samples than for vapor samples. . . more accurate estimates of air toxics emissions may be possible by predicting vapor compositions based on the measurement of liquid phase concentrations than by measuring vapor concentrations” [ 141.

The study also concluded that the determination of the liquid molecular weight (M,) is an important factor in predicting the vapor phase equilibrium concentrations. The molecular weight of the liquid is used to convert the liquid phase weight concentrations, wi, into mole concentrations, xi. These concentrations are then used in Raoult’s Law, RKS, or PR to predict vapor phase molar concentrations, y¡, as shown in Equation A- 1.

xi = wi (M,/Mi) (A- 1)

Where:

xi =molar concentration of component i in the liquid

wi = weight concentration of component i in the liquid

M/ = liquid molecular weight, (lb/lb-mole).

(mole fraction).

(weight fraction).

Table A-6-Comparison of Predicted Vapor Concentrations Using the Response Factor Analytical Data

RKS/ldeal PWldeal

No. 2 AK Spiked No. 2 AK Spiked Compounds Midcont JP-4 FO Crude AK Mean Midcont JP-4 FO Crude AK Mean

Butane 0.99 0.96 NA 0.99 0.99 0.99 0.99 0.97 NA 0.99 0.99 0.99 n-Hexane 1.01 0.96 NA I .o0 1.01 1.01 1.04 1.02 NA 1.04 1.05 i .O5 Benzene i .26 I .38 NA 1.26 1.26 1.26 1.41 1.51 NA 1.41 1.40 1.39 Cyclohexane 1 .O7 1.13 NA i .O7 1 .O7 i .O7 1.17 I .22 NA 1.17 1.16 1.16 Toluene 1.19 1.30 1.07 1.19 1.19 1.18 1.34 1.44 1.19 1.34 1.33 1.33 Octane 0.97 0.94 1.05 0.96 0.98 0.98 1.05 1.03 1.13 1 .O5 1.06 1.08

m,p-Xylenes 1 .o1 1 . 1 1 0.91 1 .O1 I .o1 I .o1 1.17 1.26 1.04 1.17 1.16 1.16 o-Xylene 1 .O8 1.20 0.95 i .O8 1 .O8 I .O8 1.26 1.37 1.09 1.26 1.25 1.25 Isopropylbenzene 0.86 0.93 0.79 0.86 0.86 0.86 1.00 1.06 0.91 1.00 0.99 0.99

0.80 0.90 0.71 0.80 0.80 0.80 0.97 1.05 0.84 0.97 0.96 0.96 Trimethylbenzene Naphthalene i .26 1.54 0.97 1.26 I .25 I .26 1.63 1.89 1.24 i .63 1.60 1.60

Mean 1.04 1.13 0.97 1.04 1.05 1.05 1.19 1.26 1.08 1.19 1.18 1.18

Ethylbenzene 1.05 1.15 0.95 1.05 1.05 1.05 1.21 1.29 1.08 1.21 1.20 1.20

I ,2,4-

Note: Ideal = Raoult’s Law; PR = Peng-Robinson equation of state; RKS = Redlich-Kwong-Soave equation of state.



--`,,-`-`,,`,,`,`,,`---

24 CHAPTER 19.4

Mi = molecular weight of component i in the liquid, (IbAb-mole).

If the molecular weight of the liquid phase is unknown, it must be determined. Two analytical methods for determining the molecular weight were used in the study: ( I ) gel permeation chromatography (GPC) using a refractive index (RI) detector, and (2) gas chromatography (GC) using a flame ionization detector (FID). The GC determination of molecular weight distribution offers better resolution of samples characterized by a low boiling point/MW distribution (boiling point <392’F or molecular weight ~ 2 5 0 ) .

The study did not determine the applicability of Raoult’s Law to mixtures containing hydrocarbon and non-hydrocarbon constituents, such as reformulated gasolines containing oxygenated species, e.g., MTBE.

A.2 Raoult’s Law Versus Alternate Methods

In the 1990 study [13], Radian Corporation also investigated alternate approaches for estimating air-toxic emissions using Henry’s Law, Petroleum K-Values, and activity coefficients. None of these approaches were found to have an advantage over Raoult’s Law. A summary of the findings using each of these methods in speciating air emissions is given below.

A.3 Henry’s Law Henry’s Law can be viewed as a general form of Raoult’s

Law and is described by Equation A-2.

yi = Hixi/P, (A-2)

Where:

Hi = Henry’s Law constant for component i (psia).

P, = total system pressure (psia).

Experience indicates that Henry’s Law can provide satisfactory results for solutes present in low concentrations in a solvent, as is the case for organic compounds present in trace concentrations in an organic liquid mixture. However, most of the Henry’s Law constants that are available in the open literature have been developed for dissolved gases in aqueous systems. Henry’s Law constants for organics that are dissolved in other organics, such as petroleum stocks and products, might not be readily available. Developing these constants would require a major research effort that may not be justified if Raoult’s Law can draw upon more readily available data and provide accurate results.

A.4 Petroleum K-Values There are available data that relate the vapor (yi) and liquid

(x i ) molar compositions as a function of temperature and pressure for a number of light organic compounds and pseudo-compounds (e.g., boiling-point cuts) found in refining applications. This relationship involves the use of K-values, or component vapor-liquid equilibrium constants, Ki, as described in Equation A-3:

yi = Kixi (A-3)

Table A-7-Comparison of Predicted Vapor Concentrations Using the Linear Regression Analytical Data

RKSAdeal PWldeal

No. 2 AK Spiked No. 2 AK Spiked Compounds Midcont JP-4 FO Crude AK Mean Midcont JP-4 FO Crude AK Mean

Butane 0.98 0.96 NA 0.98 1.04 1.00 0.99 0.97 NA 0.98 1 .O3 I .o0 n-Hexane 1.00 0.98 NA 1 .o0 I .O8 i .O3 I .O3 1 .o2 NA I .O3 1.11 I .O6 Benzene 1.28 i .38 NA 1.30 1.14 1.25 I .44 1.51 NA I .45 I .25 1.38 Cyclohexane I .O8 1.13 NA I .O8 1 .o2 1 .o6 1.19 1.22 NA 1.19 1 .O9 1.16 Tolucnc 1.21 1.29 I .O7 1.22 1.08 1.17 I .38 I .43 1.19 1.38 1.20 1.32 Octane 0.96 0.94 1.05 0.96 1.06 0.99 I .O5 1 .O3 1.13 I .o4 1.15 1 .O8 Ethylbenzene I .O7 1.14 0.95 1.07 0.96 1.04 I .24 I .29 1 .O8 1.24 1.09 1.19 m,p-Xylenes 1 .O3 1 . 1 1 0.9 i 1.04 0.92 1.00 1.20 1.25 1.04 1.21 I .O5 1.15 o-Xylene 1 . 1 1 1.20 0.95 1.11 0.97 1 .O7 1.30 1.37 1.09 1.31 1.1 1 1.24 lsopropylbenzene 0.87 0.92 0.79 NA 0.80 0.85 1 .o2 1.06 0.91 NA NA 0.98

0.82 0.89 0.71 0.83 0.72 0.79 0.99 1.05 0.84 1.00 0.85 0.95 1,2,4- Trimethylbenzene Naphthalene 1.31 1.53 0.97 1.33 1.00 1.23 1.71 1.89 1.24 1.73 NA 1.56

Mean 1.06 1.12 0.97 1.06 0.98 1.04 1.21 1.26 1.08 1.22 1.09 1.17

Note: Ideal = Raoult’s Law; PR = Peng-Robinson equation of state; RKS = Redlich-Kwong-Soave equation of state.



--`,,-`-`,,`,,`,`,,`---


Where:

K, = vapor-liquid equilibrium constant for component i (dimensionless).

These K-values have been developed principally for estimating fractionation behavior in distillation columns, and are available in the form of nomographs, or as temperature and pressure correlations. Unfortunately, the majority of specific air emissions compounds of interest are not included in the list of K-values that data have been developed for. Also, the ranges of temperature and pressure included in the correlations are typically higher than those that are likely to occur in storage tanks. A major research effort would be required to develop the necessary K-values for storage tank conditions. Therefore, using petroleum K-values to estimate speciated air emissions from storage tanks does not appear to be a practical approach at this time

AS Activity Coefficients Activity coefficients, x, may be viewed as correction fac-

tors to Raoult’s Law that take liquid-phase non-idealities into account when relating vapor-phase and liquid-phase compositions, as described in Equation A-4.

Y I = P I X , & (A-4)

Where:

3: = activity coefficient for component i (dimensionless).

Many models have been developed for estimating activity coefficients for binary mixtures. In mixtures with more than two components, activity coefficients must be obtained for each possible pair of components, and then combined using mixing rules. For complex mixtures such as refinery stocks, the effort involved to determine the constants for each binary pair would be too great to use activity coefficients as a viable method to estimate speciated air emissions from storage tanks.

A.6 Applicability of Raoult’s Law to Fuel Oxygenates

A report [I71 prepared by the University of Delaware for API contains measured vapor-liquid equilibrium data for several oxygenated fuel mixtures. The data were measured so that equations of state and activity coefficient models could

be developed to predict the shift in equilibrium pressures and vapor compositions when oxygenates, such as ethers and alcohols, are added to gasoline, thereby increasing the octane rating. Measured vapor pressures of the pure components 2,2,4-trimethylpentane (TMP), MTBE, and methylcyclohexane (MCH) are presented. Vapor-liquid equilibrium data are presented for binary mixtures MTBEíTMP and MTBW MCH. Vapor-liquid equilibrium data are also presented for MTBE in a four-component simulated gasoline mixture, MTBE in an eleven-component simulated gasoline mixture, and ethanol in an eleven-component simulated gasoline mixture. The data were taken at temperatures ranging from 20 to 140°F.

The results presented in the report represent a comprehen- sive data set that can be used to determine the effect of oxygenate additives to hydrocarbons and gasoline mixtures at temperatures around ambient conditions. The report states that for mixtures without MTBE or ethanol, the equilibrium parameters were satisfactorily described by a cubic equation of state, such as PR or RKS, without the introduction of a binary interaction parameter. With the addition of either MTBE or ethanol, a binary interaction parameter was needed between these oxygenates and the hydrocarbons. However, when this parameter was adjusted to fit the measured vapor compositions at fixed values of temperature and liquid compositions, the predicted pressure was systematically too low. The relative difference between the measured and predicted pressures increased with increased oxygenate composition and decreased with increased temperature and pressure.

In both scenarios, when the equilibrium constants, Ki, for butane and benzene were compared before and after addition of 10-20 percent methyl tert-butyl-ether (MTBE), there was less than a 10 percent variation in the equilibrium constants. However, when 8.9-10 percent ethanol was added instead of MTBE, the equilibrium constants increased for butane by less than 10 percent, and varied from a 68 percent increase at 8.9 percent ethanol addition to a 15 percent decrease at 10 percent ethanol addition. Overall, the variation of the equilibrium constants was small in relation to the expected variations associated with sample collection and analysis.

Raoult’s Law was not specifically addressed in this report; however, the variation in K values was relatively small, so Raoult’s Law does provide a reasonable estimate of the vapor composition when oxygenates, such as ethers and alcohols, are added to gasoline.



--`,,-`-`,,`,,`,`,,`---

APPENDIX B-DETERMINATION OF PHYSICAL PROPERTY DATA FOR SPECIFIC CHEMICALS AND PETROLEUM STOCKS

To calculate total hydrocarbon emissions, and specific emissions of individual hydrocarbons, the following physical properties and environmental information are required:

a. Temperature. b. Pressure. c. Molecular weight. d. Density. e. Concentration.

B.l Temperature

8.1.1 AVERAGE AMBIENTTEMPERATURE,

This is the National Weather Service's ( N W S ) average daily recorded atmospheric temperature at the closest local weather station to the facility.

8.1.2 AVERAGE DAILY TEMPERATURE CHANGE, A Ta

This is the average daily change of atmospheric temperature, as reported by the N W S . It is the difference between the daily maximum and daily minimum ambient temperatures.

8.1.3 LIQUID BULKTEMPERATURE, Tb

This is the average stock liquid bulk storage temperature. It is usually available from gauging records. If the average stock liquid bulk storage temperature, T,, is unknown, it can be estimated from the average ambient temperature, Tu, and the tank paint color by using Table B-1 .

8.2 Pressure

8.2.1 ATMOSPHERIC PRESSURE, Pa

This is the average atmospheric pressure, Pu, in pounds per square inch absolute, at the facility location. If it is not available for the site location, a value of 14.7 psia may be used.

Table B-1-Average Annual Stock Liquid Bulk Storage Temperature (Tb) as a Function of Tank Paint Color

Tank Color Tb (OF)

White To + O

GnY T, + 3.5 Black To + 5.0

Note: To = average annual ambient tempenture, in "E

Aluminum r, + 2.5

8.2.2 STOCKTRUE VAPOR PRESSURE, P

This is the true vapor pressure, P, of the stored liquid (in pounds per square inch absolute). For refined petroleum stocks and crude oil stocks, the stock true vapor pressure can be determined from the average RVP and the average liquid bulk temperature, T,, by using the appropriate vapor prèssure equation shown below. Alternatively, Figures 4 and 5 from API Publication 25 19 [4] can be used.

Equation for Stock True Vapor Pressure ( P ) of Refined Petroleum Stocks (1 psi to 20 psi RVP):

P= exp 0.7553- 413'0 ~ S l ~ g , o ( R V P ) (B-1) { [ (( T + 459.6))l

-[ 1.854 - (( :T9.6,)]So'5

-( 8742 )+15.64} (T + 459.6)

Where:

P = stock true vapor pressure (psia).

T

RVP = Reid vapor pressure (psi).

S rated ("F/vol.%).

= stock liquid temperature (OF).

= ASTM D86 distillation curve, at 10 percent evapo-

Equation for Stock True Vapor Pressure (P) of Crude Oils (2 psi to 15 psi RVP):

= exPr [(T + 2799 459.6) - 2.227 1 log ,o( R V P ) (B-2)

-( T + 7261 459.6 )+12.82}

Where:

P

T

RVP = Reid vapor pressure (psi).

= stock true vapor pressure (psia).

= stock liquid temperature (OF).

8.2.3 COMPONENT SATURATED VAPOR PRESSURE, P,"

This is the specific vapor pressure of a component in the stock liquid at the stock liquid temperature. This value can be determined from the Antoine Equation shown below. Antoine

27



--`,,-`-`,,`,,`,`,,`---

28 CHAPTER 19.4

Constants for specific components can be found in many standard physical property handbooks [5 ] . Table B-3 lists the Antoine Constants for selected hydrocarbons. The Antoine Equation for vapor pressure is shown in

(B-3) lOg/Opi = A - [B/(T + C)]

Where:

Pi =

- T -

A,B,C =

vapor pressure of component i at temperature T

temperature of the liquid (OC).

Antoine Constants.

(mmHg).

Note that the vapor pressure determined by Equation B-3 and Table B-3 will be in units of mmHg and must be converted into units of psia for use in other calculations.

If Antoine Constants for a particular component are not readily available, the specific vapor pressure can be found from published vapor pressure and temperature data similar to Table B-2. For estimating vapor pressure at a particular temperature that is not listed, the following relationship should be used:

This is the equation of the line through the points (UTl, log,,J'p,) and ( UT2, log,,Q,) on a plot of log,,#' versus 1/T. This two-point linear interpolation equation draws upon the Clap- eyron Equation (a simpler version of the Antoine Equation) and provides a method whereby the equation of the line between two adjacent data points is used to relate pressure and temperature between these points.

8.3 Molecular Weight

8.3.1 STOCK LIQUID MOLECULAR WEIGHT, M,

The molecular weight of the liquid, M,, can be determined by analysis of liquid samples, or by calculation from the liquid weight composition using Equation B-5:

M/ = l E ( W j / M i ) 03-51 Where:

Ml = stock liquid molecular weight (lb/lb-mole).

wj = weight composition of component i in liquid phase (weight fraction).

Mi = molecular weight of component i (lb/ib-mole).

8.3.2 STOCK VAPOR MOLECULAR WEIGHT, M,

The molecular weight of the vapor, M,, can be determined by analysis of vapor samples, or by calculation from the vapor molar composition using Equation B-6:

M, = z(yi Mi) (B-6)

Where:

M, = stock vapor molecular weight (lb/lb-mole).

y; = molar composition of component i in vapor phase (mole fraction).

Mi = molecular weight of component i (1bAb-mole).

In the absence of this information, a typical value of 64 lb/ lb-mole can be assumed for gasoline, and a value of 50 lb/ib- mole can be assumed for U.S. mid-continent crude oils (including both reactive and nonreactive fractions).

8.3.3 COMPONENT MOLECULAR WEIGHT, M,

The molecular weight of the individual components is needed to convert concentrations from a mole basis to a weight basis. The molecular weights of a number of compounds found in petroleum stocks are listed in Tables B-2 and B-3.

6.4 Density

8.4.1 DENSITY OF STOCK LIQUID, W,

This is usually specified by the user at the stock temperature. For gasoline, a value of 6.1 pounds per gallon can be used.

8.4.2 DENSITY OF CONDENSED STOCK VAPOR, W"

For refined petroleum stocks and crude oils, the density of the condensed vapor, W,is lower than the density of the stored liquid stock. If this information is unknown, it can be approximated from Equation B-7, which was developed primarily for gasoline:

W, = 0.08 M, (B-7)

Where:

W, = density of condensed stock vapor (Ib/gal).

M, = stock vapor molecular weight (lb/lb-mole).



--`,,-`-`,,`,,`,`,,`---


8.5 Concentration

B.5.1 COMPONENT MOLAR CONCENTRATION IN THE LIQUID, X,

When needed to determine vapor emissions, the molar concentration of specific chemical constituents in the petroleum stocks handled at a source can be determined from a representative summary of samples collected from the tank, or from stocks used to make up the petroleum blend.

There are two ways of calculating the molar composition of the liquid when the concentration is given on a weight basis. If the average molecular weight of the stock, M,, is known, then the molar composition can be determined from the following relationship:

x; = w;(M,/M;) (B-8)

Where:


(weight fraction). wi = weight concentration of component i in the liquid

M, = molecular weight of the liquid (IbAb-mole).

Mi = molecular weight of component i in the liquid (IbAb- mole).

For example, octane (Mi = I14 IbAb-mole) is found to represent 1 percent (weight basis) of a gasoline mixture having an average liquid molecular weight of 200 IbAb-mole. The mole fraction, xi, is:

xi = 0.01 (200/114) = 0.02 IbAb-mole

The alternate method is to:

a. Assume a unit volume of liquid (e.g., assume 1 ml if the concentrations are given in terms of mass/ml, or assume 1 liter if the concentrations are given in terms of massAiter). b. Multiply each liquid component concentration by this unit volume (e.g., 1 mi) to yield the mass for each component. c. Divide the mass of each component in the 1 ml sample by its molecular weight. This yields the molar concentration of each component.

d. Sum the moles of all the components. e. If some of the heavies in the stock are unknown, the mis- sing components can be assumed to have an average molecular weight representative of the heavier components. A typical value of 150 IbAb-mole has been used for JP-4 aviation fuel to represent the molecular weight of the heavies in the stock [ 141. Similarly, a 200 IbAb-mole has been used for No. 2 fuel oil and crude oils [14]. f. Divide the moles of a particular component by the total moles. This yields the mole fraction of that particular component.

8.5.2 COMPONENT MOLAR CONCENTRATION IN THEVAPOR, Y,

When using an equation of state to determine the vapor composition, the molar concentration in the vapor, y;, is used. To determine the mass quantity of hydrocarbon emissions represented by a specific constituent, the molar concentration, y;, needs to be converted to weight concentration, z;, using Equation B-9:

z; = y; (M;/MJ (B-9)

Where:

zi = weight concentration of component i in the vapor

=molar concentration of component i in the vapor

phase (weight fraction).

phase (mole fraction). yi

Mi = molecular weight of component i in the vapor phase (IbA b-mole) .

mole). M, = average molecular weight of the vapor phase (IbAb-

If M, is unknown, it can be determined from Equation B-6.

M" = c(y; M;) (B-10)

For example, it has been determined that octane represents 3 percent (molar basis) of the total vapor composition from a tank. The average molecular weight of the vapor is 60 IbAb- mole. The weight fraction of octane present in the vapor composition, z;, would be:

zj = 0.03 (1 14/60) = 0.06 mole fraction.



--`,,-`-`,,`,,`,`,,`---

30 CHAPTER 19.4

Table BP-Vapor Pressures of Selected Compounds

Normal Liquid Boiling Point Density Vapor Pressure (psia)

Compound (OF) (Ib/ga@ 60°F) (40’F) ( 5 O O F ) (60°F) (70°F) (80’F) (90°F) (100*F) 133.3 6.652 1.714 2.255 2.932 3.768 4.792 6.033 7.525 Acetone

Acetonitrile Acrylonitrile Allyl alcohol Allyl chloride Ammonium hydroxide Benzene

tert-butyl alcohol n-butyl chloride Carbon disulfide Carbon tetrachloride Chloroform Chloroprene Cyclohexane Cyclopentane 1.1 -Dichloroethane 1.2-Dichloroethane cis- 1,2-Dichloroethyle Diethylamine Diethyl ether Di-¡so-propyl ether I &Dioxane Di-n-propyl ether Ethyl acetate Ethyl acrylate Ethyl alcohol Freon 1 1 n - H e p t an e n-Hexane Hydrogen cyanide “isooctane” ¡so-Pentane Isoprene iso-propyl a:cohd Methacrylonitrile Methyl acetate Methyl acrylate Methyl alcohol Methylcyclohexane Methy Icyclopentane Methylene chloride Methyl ethyl ketone Methyl methacrylate Methyl propyl ether Nitromethane n-Pentane n-Propylamine 1,1, I -Trichloroethane Trichloroeythlane Toluene Vinyl acetate Vinylidene chloride

isû-butyl alcohol

178.8 17 1.2 206.7 113.2 28.1 176.1 225.8 180.3 173.1 115.2 169.9 142.1 138.9 177.3 120.6 135.1 182.2 140.9 131.8 93.9 154.9 214.3 194.1 170.7 211.1 172.9 74.9 209. I 155.7 78.2 210.6 82. I 93.3 180.0 194.5 134.5 176.3 148.4 213.7 161.2 103.5 175.3 212.5 102.3 214.1 96.9 118.1 165.3 188.5 231.1 162.5 88.8

6.564 6.774 7.147 7.870 6.942 7.363 6.721 6.528

10.600 13.360 12.520 8.032 6.521 6.253 9.868 10.520 10.770 5.943 5.999 6.094 8.669 6.265 7.549 7.729 6.627 12.520 5.753 5.544 5.788 5.762 5.222 5.722 6.593 6.7 15 7.845 8.002 6.663 6.459 6.285

11.150 6.757 7.917 6.122 9.523 5.266 6.036

I 1.220 12.290 7.280 7.824 10.200

-

0.643 0.776 0.128 2.964 19.210

0.039

0.703 2.974 0.820 1.443 1.748 0.780 2.554 1.705 0.53 1 1.518 1.762 4.426 1.092

0.405 0.605 0.227 0.316 7.061 0.288 1.108 6.236 0.326 6.061 4.67 I 0.226 0.504 1.531 0.582 0.758 0.299 1.007 3.463 0.625 0.217 3.557 0.219 4.331 2.285 0.900 0.507 O. 175 0.761 5.059

-

-

-

0.858 1 .O33 O. 188 3.806 24.060 0.878 0.063

0.944 3.790 1 .O90 1.907 2.266 0.994 3.285 2.236 0.722 1.998 2.314 5.660 1.441

0.554 0.828 0.317 0.458 8.832 0.397 1.468 7.909 0.444 7.598 5.918 0.334 0.672 2.039 0.790 1.061 0.409 1.33 1 4.482 0.845 0.307 4.593 0.305 5.495 3.014 1.195 0.68 1 0.243 I .O3 I 6.426

-

-

1.131 1.357 0.272 4.832 29.840 1.166 0.099

1.25 1 4.781 1.430 2.486 2.903 1.230 4.180 2.899 0.967 2.597 3.002 7.157 1.879 0.452 0.747 1.117 0.436 0.653 10.940 0.540 1.919 9.935 0.597 9.428 7.419 0.485 0.885 2.680 I .O58 1.462 0.552 1.737 5.730 1.127 0.426 5.865 0.419 6.898 3.927 1.566 0.905 0.333 1.378 8.068

-

1.414 1.761 0.388 6.071 36.670 1.529 0.152

I .637 5.973 1.854 3.203 3.680 1.601 5.263 3.716 1.279 3.335 3.853 8.953 2.423 0.606 0.994 i ,487 0.591 0.916 13.420 0.723 2.477 12.370 0.791

11.590 9.208 0.692 1.153 3.480 1.398 1.988 0.734 2.241 7.242 I .484 0.583 7.409 0.568 8.573 5.058 2.028 1.187 0.451 1.819 10.020

-

1.901 2.260 0.544 7.552 44.680 1.980 0.228 0.910 2.1 17 7.395 2.376 4.078 4.616 2.060 6.562 4.7 13 1.669 4.237 4.892

1 1 .O90 3.093 0.804 1.306 1.955 0.792 1.267 16.320 0.956 3.161 15.270

1 .O35 14.120 1 1.320 0.973 1.484 4.468 1.824 2.666 0.964 2.858 9.055 1.932 0.786 9.266 0.761 10.560 6.445 2.596 1.541 0.601 2.373 12.330

2.425 2.870 0.753 9.307 53.980 2.537 0.337 1.282 2.707 9.079 3.012 5.137 5.735 2.620 8.105 5.9 18 2.154 5.327 6.151 13.610 3.910 1 .O54 1.696 2.540 1 .O48 1.728 19.690 1.249 3.990 18.690 1.339 17.070 13.800

I .347 1.891 5.674 2.355 3.533 1.25 I 3.607 11.210 2.487 1 .o45 11.480 1.005 12.880 8.127 3.289 1.978 0.792 3.062 15.020

3.066 3.610 1.028 11.370 64.720 3.216 0.487 1.776 3.425

11.060 3.781 6.407 7.061 3.300 9.924 7.363 2.750 6.632 7.662 16.550 4.898 1.367 2.178 3.263 1.371 2.328 23.580 1.614 4.985 22.720 1.7 14 20.470 16.690 1.839 2.386 7.133 3.008 4.628 1.606 4.507 13.750 3.169 1.373 14.100 1.314 15.590 10.150 4. I24 2.515 1 .O32 3.908 18.140



--`,,-`-`,,`,,`,`,,`---


Table B-3-Physical Properties and Antoine Constants for Selected Hydrocarbons

Name Formula Molecular Boiling Point at I Liquid Density at Antoine Constants'

Weight Atmosphere, O F 60°F. ppg A B C Acetone C H , C OC H 58.08 133.0 6.628 7.11714 1210.595 229.664 Acetonitrile Acrylonitrile Allyl alcohol Allyl chloride Ammonium hydroxide

(28.8% solution) Benzene

Tea-butyl alcohol n-Butyl chloride Carbon disulfide Carbon tetrachloride Chloroform Chloroprene Cyclohexane Cyclopentane 1.1-Dichloroethane i .2-Dichloroethane Cis-I ,2-dichloroethylene Trans- I ,2-dichloroethylene Diethylamine Diethyl ether Di-iso-propyl ether 1 ,CDioxane Dipropyl ether Ethyl acetate Ethyl acrylate Ethyl alcohol Freon 11 n-Heptane n-Hexane Hydrogen cyanide Isoctane Isopentane Isoprene Isopropyl alcohol Methacrylonitrile Methyl acetate Methyl acIylate Methyl alcohol Methylcyclohexane Methylcyclopentane Methylene chloride Methyl ethyl ketone Methyl methacrylate Methyl propyl ether Nitromethane n-Pentane n-Propylamine l,l,l-Trichlorethane Trichloroethylene Toluene Vinyl acetate Vinylidene chloride

Iso-butyl alcohol

41.05 53.06 58.08 76.53

35.05

78.11 74.12 74.12 92.57 76.13

153.84 119.39 88.54 84.16 70.13 98.97 98.97 96.95 96.95 73.14 74.12

102.17 88.10

102.17 88.10

100. 1 I 46.07

137.38 100.20 86.17 27.03

114.22 72.15 68.1 1 60.09 67.09 74.08 86.09 32.04 98.18 84.16 84.94 72.10

100.1 1 74.12 61.04 72.15 59.1 1

133.42 131.40 92.13 86.09 96.5

178.9 173.5 206.6 113.2

83.0

176.2 227.1 180.5 172.0 115.3 170.2 142.7 138.9 177.3 120.7 135.1 182.5 140.2 119.1 131.9 94.3

153.5 214.7 195.8 170.9 21 1.8 173.1 75.4

209.2 155.7 78.3

2 10.6 82.1 93.5

180.1 194.5 134.8 176.9 148.4 213.7 161.3 104.2 175.3 212.0 102.1 214.2 96.9

119.7 165.2 188.6 231.1 162.5 89.1

6.558 6.758 7.125 7.864

7.481

7.365 6.712 6.595 7.430

10.588 13.366 12.488 8.046 6.522 6.248 9.861

10.500 10.763 10.524 5.906 5.988 6.075 8.659 6.260 7.55 1 7.550 6.610

12.480 5.727 5.527 5.772 5.794 5.199 5.707 6.573 6.738 7.83 1 7.996 6.630 6.441 6.274

11.122 6.747 7.909 6.166 9.538 5.253 6.030

11.216 12.272 7.261 7.817

10.383

7.1 1988 7.03855

5.29716 11.1870

-

6.90565 7.47680 7.4743 1 6.83694

6.87926 6.4943 6.16 150 6.84130 6.88676 6.9770 7.0253 7.0223 6.965 1 5.8016 6.92032 6.8495 7.43155 6.9476 7.10179

8.32109 6.88428 6.89677 6.87601

6.8 1 189 6.83315 7.01 187 8.1 1778 6.9802 7.0652

4.89750 6.82300 6.86283 7.4092 7.06356 8.4092 6.1186 7.28166 6.852% 6.92651 8.6434 6.5183 6.95464 7.2101

-

-

-

-

-

1314.4 1232.53 4068.5 418.375

- 121 1.033 1362.39 1314.19 1173.79

121 2.021 929.44 783.45

1201.53 1 124.162 1174.02 1271.3 1205.4 1141.9 583.30

1064.07 1139.34 1554.68 1256.5 1244.95

1718.10 1043.004 1264.90 1171.17

1257.84 1040.73 1126.159 1580.92 1274.96 1 157.64

1474.08 1270.763 1186.059 1325.9 126 1.34 2050.5 708.69

1446.94 1064.84 1044.05 2136.6 1018.6 1344.800 1296.13

-

-

-

-

-

230.0 222.47 392.7 128.168

-

2 2 O. 7 9 O 178.77 186.55 218.13

226.41 196.03 179.7 222.65 231.36 229.06 222.9 230.6 231.9 144.1 228.80 218.7 240.34 219.0 217.88

237.52 236.88 2 16.54 224.41

220.74 235.45 238.88 219.61 220.7 219.73

229.13 22 i .42 226.04 252.6 22 1.97 274.4 179.9 227.60 233.01 210.84 302.8 192.7 219.48 226.66

-

-

-

-

-

Note: Most of the values in this table were taken or calculated from data in References 22 and 23. Antoine Constants from Reference 5. aThe Antoine Constants listed in this table are for use in the Antoine Equation, Log,, P = A-[B/(T+C)], where T is the temperature in "C and P is the vapor pressure in mmHg.



--`,,-`-`,,`,,`,`,,`---

APPENDIX C-COMPARISON OF MOLECULAR WEIGHT, BOILING POINTS, VAPOR PRESSURE, AND BLENDING RVP FOR SELECTED HYDROCARBONS AND OXYGENATES

Pure substances that have strong inter-molecular attrac- tions boil at temperatures far higher than those suggested by their molecular weight when compared to substances without these bonding tendencies. A prime example of strong hydrogen bonding is water, which has a molecular weight of 18, yet boils at 212°F at atmospheric pressure. Methane has a similar molecular weight of 16, but boils at -259°F. When pure substances with strong bonding tendencies, such as water or alcohols, are diluted by other substances, they will appear to have much higher vapor pressures in the mixture than they do as pure substances.

Table C-1 contains a number of chemical substances that might be found as constituents in crude oil fractions, petroleum refinery naphtha and distillate streams, or blended petroleum products such as gasoline. Those exhibiting high boiling points and low vapor pressures as pure substances for their relative molecular weights are highlighted, such as water and the alcohols, and to a much lesser extent, aromatics and cycloparaffins. These same data are plotted for boiling points in Figure C-1, and for vapor pressure in Figure C-2. In general, at any molecular weight range, isoparaffins have the lowest boiling point and highest vapor pressure. They are fol- lowed by normal paraffins, cycloparaffins, aromatics, and alcohols (keyed as XOH in the chart legend). Ethers such as MTBE fall into the range of the iso- and normal paraffins. (Ethers are keyed as XOX in the chart legend.)

The last column of Table C-1 presents RVP blending values that might be used as TVP values at 100°F in speciating storage tank vapors from liquids containing oxygenates. Note that the blending RVP for MTBE is essentially the same as its pure substance, TVP at 100°F. If available, companies might want to use their own blending values instead of the values provided in this material.

Table C-1-Chemical Substances Found as Constituents in Crude Oil Fractions

NBP

VP @ lûû°F Compound (OD @ia)

Methane -259 Water Ni trogen Ethane Methanol Hydrogen sulfide Carbon dioxide Propane Ethanol Isobutane n-Butane Isopropanol Cyclopentane Isopentane n-Pentrane Methyl-n-propyl ether 2-Methyl-2-propanol Benzene Methylcyclopentane Cyclohexane 2,3-Dimethylbutane 2-Methylpentme n-Hexane Methyl ten-butyl ether Toluene Methylcyclohexane 2-Methylhexane n-Heptane Ethylbenzene m-Xylene I ,2,4-Tnmethylpentane 2-Methylheptane n-Octane I ,2,4-Trimethylbenzene Cumene n-Nonane n-Decane n-Undecane

212 -32 1 -128 I48 -77

-44 173 I l 31

180 121 82 97

1 02 I80 176 161 177 i 36 I40 i 56 131 23 I 214 194 209 277 282 21 I 244 258 337 306 303 345 385

1 .o

704.0 4.6

395.0 1072.0 189.0

2.3 72.8 51.7

1.8 9.9

20.5 15.6 14.1 1.8 3.2 4.5 3.3 7.4 6.8 5.0 8.0 1 .o I .6 2.3 I .6 0.4 0.3 1.7 0.8 0.5 O. 1 0.2 0.2 o. I 0.02

33



--`,,-`-`,,`,,`,`,,`---

34 CHAPTER 19.4

400

300

200

1 O0

!+

a 3 E 2

i c .-

o> C = o -

-1 O0

-200

300

400

A

Molecular Weight

Gas i-Paraf A n-Paraf X Cvclo X Arom O XOH + XOX I Legend: I

Figure C-1-Normal Boiling Point Versus Molecular Weight



--`,,-`-`,,`,,`,`,,`---


m u> .- o. o-' 4

f

O

c m

m m 9 a

8 >

t I I

1 O0

* A- X A

X e * X b +

0.1 I

0.01 I I

Molecular Weight

I ¡-Paraf W n-Paraf A Cycio X Arom X XOH O XOX 1 Legend:

Figure C-2-Pure Substance Vapor Pressure at 100°F Versus Molecular Weight



--`,,-`-`,,`,,`,`,,`---

APPENDIX D-REFERENCES

1.

2.

3.

4.

5 .

6.

7.

8.

9.

American Petroleum Institute, Atmospheric Hydrocarbon Emissions From Manne Vessel Transfer Operations, API Publication 25 14A, Second Edition, September 1981.

American Petroleum Institute, Evaporative Loss From External Floating-Roof Tanks, API Publication 25 17, Third Edition, February 1989.

American Petroleum Institute, MPMS 19.1, Evaporative Loss From Fixed-Roof Tanks, Second Edition, API Bulle- tin 2518, October 1991.

American Petroleum Institute, Evaporation Loss From Internal Floating-Roof Tanks, API Publication 25 19, Third Edition, June 1983.

Dean, J. A. Ed., h n g e S Handbook of Chemistry, Twelfth Edition, McGraw-Hill Book Company, New York, 1979.

Felder, R. M. and Rousseau, R. W., Elementary Princi- ples of Chemical Processes, John Wiley and Sons, New York, 1978.

Gallant, R. W., Physical Properties of Hydrocarbons, Vol- ume l , Gulf Publishing Company, Houston, Texas, 1974.

Mass, J. M., “Continuous Distillation: Separation of Binary Mixtures,” Section 1.1 of Handbook of Separation Techniques for Chemical Engineers, P. A. Schweitzer, editor-in-chief, McGraw-Hill Book Company, New York, 1979.

Murphy, P., Midwest Research Institute, Estimating Air

prepared for the American Petroleum Institute, Septem- ber 1990.

14. Skinner, F. D., D. A. Brymer, and P. A. Thompson, Radian Corporation, Analysis of Crude Oil and Rejned Product Samples and Comparison of Vapor Composition to Model Predictions, Final Report, prepared for the American Petroleum Institute, June 1992.

15. U. S . Environmental Protection Agency, Office of Air and Radiation, Office of Air Quality Planning and Standards, Compilation of Air Pollutant Emission Factors, Fourth Edition, AP-42, Chapter 12, “Storage of Organic Liquids,” Research Triangle Park, North Carolina, October 1992.

I

16. Wright, D. A., M. N. Menon, and S. H. Peoples, AB2588 Emission Estimation Techniques for Petroleum Refiner- ies, Draft Final Report, prepared for Western State Petroleum Association, July 12, 1989.

17. Wu, H. S., K. A. Pividal, and S. I. Sandier, “Vapor-Liquid Equilibria of Hydrocarbons and Fuel Oxygenates,” J. Chem. Eng. Data, 1991, Vol. 36, No. 4, pp 41 8-421.

18. Weast, R. C., Ed., Handbook of Chemistry and Physics, 57th Edition, The Chemical Rubber Co., Cleveland, Ohio, 1976-1977. Refer also to Chapter 5 of the API “Technical Data Book-Petroleum Rejning. ” 5th Edition, 1992.

19. Soave, G., “Equilibrium Constants From a Modified Redlich-Kwong Equation of State,” Chem. Eng. Sei., 1972, Vol. 27, No. 6, p. 1197.

Tox& ~ Emissions from Organic Liquid Storage Tanks, EPA Report No. EPA-450/4-88/004, October 1988. 20. Peng, D.Y. and Robinson, D.B., “A Two Constant Equation

of State,” I.E.C. Fundamentals, 1976, Vol. 15, pp 59-64.

10. Null, H. R., Phase Equilibrium in Process Design, Wiley- Interscience, New York, 1970. 21. ASPEN/SP Process Simulator, Simulation Sciences, Inc.,

Houston, Texas. 11. Reid, R. C., J. M. Prausnitz, and B. E. Poling, The Pmp-

erties of Gases and Liquids, Foufi Edition, McGraw- Hill Book Company, New York, 1987.

22. Timmermanns, J., Physio-Chemical Constants Of PUR Organic Compounds, Elsevier, New York, 1950.

12. Robinson, R. N., Chemical Engineering Reference Man- 23. Perry, R.H., C.H. Chilton and S.D. Kirkpatrick, eds., ual, Fourth Edition, Professional Publications, Inc., Chemical Engineers Handbook, Fourih Edition, Belmont, CA, 1987. McGraw-Hill, New York, 1963.

13. Skinner, F. D., Radian Corporation, Review of Air Toxics Emission Calculations from Storage Tanks, Final Report,

24. American Petroleum Institute, MPMS 19.2, Evaporative Loss From Floating Roof Tanks, April 1997.

37



--`,,-`-`,,`,,`,`,,`---

Date:

Invoice TO - O Check here if same as “Ship To”

Company Company

~ame/Deot. Namdüeot.

I (Monlh, &y, Year) Ship TO - (UPS will not deliver to a P.O. Box)

Address Address

City StateRrovince City StateProvince

zip Countrv Zio Countrv

Customer Daytime Telephone No.

Fax No. Fax No

Customer Daytime Telephone No.

~~

I PREPAID AND CREDIT CARD ORDERS AR€ NOT CMRGED FOR AND HANDLING TO US. AND CANADIAN DESTINATIONS I - ~ ~~ ~~~

O Payment Enclosed $

Ci Payment By Charge Account: O MasterCard O Visa O American Express

Account No.

Name (As it m ~ e a r ~ on Card)

Emiration Date

O Please Bill Me PO. No.

~ - ~

Customer Account No.

State Sales lax - The American Petnileum Instiiutt! is requiml ur uillea sales tax 1111 publications mailed Ur the following states: AL AR, CT DC, FI, CA, II, IN, IA, U, Ky, ME, MD, ;M4 MI, MN, YO, NE, NJ, Ny, NC, ND, OH. 4 KI, Sc, TN, TX. Vr, Vh, W, and Wl. Prepayment d orders shiippd io thw siaia should include applicahle sales tax unles a purchaser is exempt. If exempt, please print yiiur statt! exrmptiirn number and enclw a uipy (if the cunent exernpticin certificate.

Exemptiirn Numkr Stak

Shipping and Handling - All orders are shipped via UPS or First Class Mail in the US. and Canada. Orders to all other countries will be sent by Airmail.

Rush Shipping Charge - Federal Express, $10 in addition to customer providing Federal Express

Subtotal

State Sales Tax (see above)

Rush Bulk Orders - 1-9 items, $10. Over 9 items, add $i each forevery additional item. Non; sh@ing on fore@ orders cannot be rusbeà without Federal E @ w acwunl number Bill and Ship Orders - U.S. and Canada, $4 per order handling fee, plus actual shipping costs. All other countries, for Airmail (standard service) add 25% of order value. All other countries, for UPS Next Day, add an additional 10% of order value.

Shipping and Handling (see 18) Total (in US. Dolkrrs)

Mail Orders: American Petroleum Institute, Order Desk, 1220 L Street, N.W., Washington, DC 200054070 Fax Orders: (202) 9624776 Phone Orders: (202) 682-8375

To better serve you, please refer to this code when ordering: [HI [SI 141 El 151 ml E] @I 171 Copyright American Petroleum Institute Provided by IHS under license with API


--`,,-`-`,,`,,`,`,,`---

The American Petroleum Institute provides additional resources and programs to industry which are based on API Standards. For more information, contact:

Training and Seminars

Inspector Certification Programs

American Petroleum Institute Quality Registrar

Monogram Licensing Program

Engine Oil Licensing and Certification System

Petroleum Test Laboratory Accreditation Program

Ph: 202-682-8490 Fax: 202-682-8222

Ph: 202-682-8161 Fax: 202-962-4739

Ph: 202-962-4791 Fax: 202-682-8070

Ph: 202-962-4731 Fax: 202-682-8070

Ph: 202-682-8233 FU: 202-962-4739

Ph: 202-682-8064 Fax: 202-962-4739

In addition, petroleum industry technical, patent, and business information is available online through API EnCompass'". Call 2 12-366-4040 or fax 2 12-366-4298 to discover more.

To obtain a free copy of the API Publications, Programs, and Services Catalog, call 202-682-8375 or fax your request to 202-962-4776. Or see the online interactive version of the catalog on our World Wide Web site - http://www. apí .org.


Helping You Get The Job Done Right.



--`,,-`-`,,`,,`,`,,`---



--`,,-`-`,,`,,`,`,,`---

Additional copies available from API Publications and Distribution: (202) 682-8375

Information about API Publications, Programs and Services is available on the World Wide Web at: http://www.api.org

American 1220 L Street, Northwest Petroleum Washington, D.C. 20005-4070 Institute 202-682-8000 Order No. H1904



--`,,-`-`,,`,,`,`,,`---

Documents

API MPMS C19.4 97